(Initiative For Policy Dialogue Series C) Joseph E. Stiglitz, Andrew Charlton-Fair Trade For All - How Trade Can Promote Development-Oxford University Press (2006) PDF

Fair Trade For All
This page intentionally left blank

Fair Trade
For All
3
how trade can promote
development
Joseph E. Stiglitz and Andrew Charlton
1
3
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Cape Town Dar es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States
by Oxford University Press Inc., New York
© Oxford University Press, 2005
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
First published 2005
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the appropriate
reprographics rights organization. Enquiries concerning reproduction
outside the scope of the above should be sent to the Rights Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose the same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Data available
Typeset by Newgen Imaging Systems (P) Ltd., Chennai, India
Printed in Great Britain
on acid-free paper by
Clays Ltd., St Ives plc., Suffolk
ISBN 0–19–929090–3 978–0–19–929090–1
1 3 5 7 9 10 8 6 4 2
Foreword
3
By the end of the twentieth century trade liberalization had become
part of the mantra of political leadership of both the left and the right
in the advanced industrial countries. President Clinton had hoped
that a new round of trade negotiations, which was to have been
launched at the Seattle meeting of the WTO in December 1999,
would be his final achievement in helping create a new world of
trade liberalization, capping the successful creation of NAFTA and
the completion of the Uruguay Round. Perhaps the new round would
be remembered as the Seattle Round, or even better, as the Clinton
Round, as previous rounds had been named after the city where they
were started (e.g. the Torquay Round of 1951 and the Tokyo Round
of 1973–9) or the official who came to be identified with the talks
(e.g. the Dillon Round of 1960–1 and the Kennedy Round of 1964–7).
As Chief Economist of the World Bank, I was greatly worried
about the imbalances of the Uruguay Round, and sensitive to the fact
that it had not delivered on the promises that had been made to the
developing countries. In an address to the WTO in Geneva, I docu-
mented those imbalances and called for a Development Round to
redress them.1 Just days before the WTO meeting convened in
Seattle (in an address at Harvard University) I predicted that unless
redressing those imbalances was at the top of the agenda, the develop-
ing countries would reject another round of trade negotiations. As it
turned out, Seattle was a watershed. The riots and protests on the
streets during the conference were the most public manifestation of
a shift in the debate about trade and trade liberalization—and of a
more significant shift in the relationship between the developed and
the developing world.
1
See Stiglitz (1999b, c).
vi FOREWORD
At the turn of the millennium, there was a new sense of collective

responsibility for the challenges faced by poor countries, and also a
recognition of the inequities created by previous rounds of trade
negotiations. The advanced industrial countries responded to the
events at Seattle and the broader public support for a new approach
to international issues. At Doha, in November 2001, they agreed to
an agenda that they claimed reflected the concerns of the developing
countries.
But a year and a half later it was clear that the developed countries
were, by and large, reneging on the promises they had made at Doha.
It appeared that even if progress were made in agriculture, it would
be slow—it might even be years before subsidies were cut back to the
1994 levels. Until just before the meeting in Cancún, in September
2003, the United States was the only country still holding out on an
agreement to make life-saving medicines available, but even after it
caved in to pressure it appeared as if it were demanding severe
restrictions on the availability of such medicines. The terms it was
forcing on developing countries—and even on Australia—in bilateral
agreements made clear that there was no intent to make it easy for
countries to have, say, generic drugs at affordable prices.
Not one of the trade ministers of the developed countries will defend
the inequities of, say, the agricultural provisions. When an earlier
version of this report was presented at the UN, at the invitation of
the President of the General Assembly, and when it was presented at
the WTO in Geneva, no one, not even the representatives of the
United States, challenged the charges that we made against the gross
inequities of the previous rounds, or even the inequities of some
of the proposals then under discussion. But the trade ministers say
in private, ‘What are we to do? Our congresses and parliaments
have tied our hands. We cannot tame the special interests. We live in
democracies, and that is part of the price one has to pay for democ-
racy. We are doing the best we can.’
At Cancún, for perhaps the first time, there was sufficient trans-
parency that journalists could cover what was going on. There were
quick reports back to national capitals about daily developments
in the negotiations. In effect, the democracies of the developing
countries replied: ‘We too live in democracies. Our democracies are
FOREWORD vii
demanding that we sign a fair agreement. If we return with another

agreement as unfair as the last, we will be voted out of office. We too
have no choice.’ So the choice for the world was between a fair
agreement reflecting the sentiments of a broad majority of the
populations in both developed and less developed countries, or
another unfair agreement, reflecting the special interests in devel-
oped countries. It was clear that the developing countries were on
far higher moral ground than were the developed countries.
In the aftermath of the failure of Cancún, the Commonwealth
countries—a group of nations with a historical association to the
United Kingdom—asked us to undertake a study of the Development
Round. The 52 Commonwealth countries consist of developed
countries (the UK, Canada, Australia, South Africa, and New Zealand),
and large developing countries (India, Pakistan, Nigeria, Malaysia)
and many small countries (including Saint Kitts, Fiji, Cyprus). Thus
the Commonwealth provides a unique forum in which the vital
issues affecting the relationship between developing and developed
countries can be discussed in a spirit of openness and understanding.
The Commonwealth posed the question: ‘What would a true
development round—one reflecting the interests and concerns of the
developing world, one which would promote their development—
look like?’ Our answer was that it would look very different from
that embodied in the agenda that was set forth in Doha, and even
more different from how matters had evolved subsequently. We
came to the conclusion that the so-called ‘Development Round’ did
not deserve its epithet. This book puts the recommendations of
that report within the broader context of trade policy, development,
and the WTO.
There are some people that will criticize the content and motivation
of this book. There is certainly a concern that by pointing out the
unfairness of global trade rules, this book will cause governments
and vested interests in developing countries to blame outsiders for
their problems rather than engage in difficult internal reform. But
like the result of any analysis, information can be misused, and the
only protection is to be as clear as one can about the assumptions
underlying the analysis. While it is true that developing countries
could do more for themselves, and that many of their problems are
viii FOREWORD
only marginally related to constraints on external market access,

that is no excuse for an international trade regime that makes life
more difficult for the developing countries. The fact that the truth
might lead individuals to unhappiness as they realize how poorly
they have been treated can hardly be an argument for not engaging in
analysis and disseminating results. There is, of course, the risk that
recalcitrance in the North and unrealistic expectations in the South
could lead to a stalemate. But this book, by showing that there is in
fact a rich agenda ahead, provides a variety of channels through
which progress can occur.
Most of the book is an incidence analysis. It describes the policies
that would do the most to integrate the developing countries into the
world trading system, to give them new trading opportunities, and to
help them to capitalize on those opportunities. It is premised on the
hope that a better understanding of the effects of trade agreements
will help mobilize public opinion in both developed and less devel-
oped countries; that it will strengthen the case for negotiators in the
hard bargaining that marks any round of trade negotiations; and
that it will help bring about reforms in the procedures and in the
institutions of the WTO which will enhance transparency and more
equitable outcomes. As the old aphorism has it, knowledge is power.
It is our hope that the information provided by this book might play
a small role in shaping the outcome of trade negotiations.
We should clarify what this book is, and what it is not. It is a review
of the theoretical and empirical evidence—much of the detail of
which is located in the Appendices—concerning the impact of pro-
visions of previous trade agreements and proposed new agreements
on the well-being and development of the developing countries. On
the basis of that review we delineate a set of priorities for a ‘true’
Development Round. The book itself does not undertake any original
analyses of these impacts, though we comment on the assumptions,
strengths, and weaknesses of studies already in the literature.
If there is a successful conclusion to the Doha Round—or to any
subsequent round of trade discussions—developing countries will
need substantial assistance to enable them to adapt to the resulting
changes, and to take advantage of any resulting new opportunities.
Thus, the second question we address is: ‘What kind of assistance
FOREWORD ix
should be provided by the advanced industrial countries?’ But before

addressing that question, we needed to ask prior ones: ‘Why is such
assistance so important? Why are the costs of adjustment for devel-
oping countries higher, and their ability to bear those costs so much
lower, than for developed countries?’ It is our hope that by making it
clear why assistance is so important if trade liberalization is to bring
its potential benefits to developing countries, we can further
increase the resolve of developed countries to live up to the commit-
ments they have already made to provide additional assistance to the
developing countries. Just as the developed countries appear to have
fallen markedly short of their commitments to the developing coun-
tries and to each other that they made at Doha in November 2001, to
make the current round of trade negotiations a Development Round,
at least some of the developed countries appear to have fallen
markedly short of the commitments in financial assistance that they
made at Monterrey in March 2003. These were commitments based
on the noblest of ambitions, to create a fairer globalization and to
increase the well-being of the world’s poorest. It is our hope that this
book may, in some small way, contribute to the achievement of
these ambitions.
Joseph Stiglitz
2005
Preface
3
This book goes to press as the world moves towards the WTO’s 6th
Ministerial Meeting in Hong Kong in December 2005. This is the
first Ministerial Meeting since the 5th Ministerial in Cancun,
Mexico, collapsed in failure and recriminations in 2003. Progress
since Cancun on the central issues in dispute, including agriculture,
has been slow, and there has been growing pessimism about the
potential outcomes of the Hong Kong negotiations. The optimists
hope not only for an agreement, but one which is more than just a
face-saving gesture, such as a pro-forma commitment to continue
discussions and a reiteration of the lofty goals set at Doha.
The document which launched this round of talks—The Doha
Declaration—was full of noble ambition. It promised to rectify the
imbalances of previous rounds of trade agreements, that had left, for
instance, developed country tariffs against developing country
products far higher than those against developed countries. The
world has come to recognize the imperative of reducing poverty in
the Third World. It has agreed upon a set of targets—the Millennium
Development Goals. And it has increasingly come to recognize
the importance of opening up trade opportunities for the develop-
ing countries—and providing them assistance to grasp these
opportunities—if these targets are to be met. It was accordingly
totally appropriate that at Doha the trade ministers agreed to
make the Round of trade negotiations they were then launching a
development round, one which would help, not hinder, the develop-
ing countries in achieving those aspirations. The rest of the world
cannot solve the problems facing developing countries—their suc-
cess will depend largely on their own efforts—but it should not tilt
the playing field against them, which, as we have shown, in many
respects, it has been doing.
xii PREFACE
Part of the problem has been that discussions on reforming the

global trading system have been approached as a pure matter of
bargaining—and in the bargaining, the poor and the weak, the devel-
oping countries, almost inevitably come out short. Even had the
agenda that had been set out in Doha been more fully accomplished,
it would have been a far cry from a true development agenda. But
what has been emerging since then clearly does not deserve that epi-
thet. The irony is that both the North and the South as a whole could
benefit from a fair and development-oriented agenda.
This book has made it clear that, regardless of the outcome of
Hong Kong, we have a long way to go if we are to establish a global
trading regime which represents fair trade for all. We should, how-
ever, be content with nothing less.
Acknowledgements
3
This book has been written by Joseph Stiglitz and Andrew Charlton,
on behalf of the Initiative for Policy Dialogue (IPD), a network of
some two hundred economists and development researchers
throughout the developed and developing world who are committed
to furthering the understanding of the development process and
policies that would enhance development. This project was managed
by the IPD Managing Director, Shari Spiegel, and assisted by the
IPD Program Coordinator, Shana Hofstetter, and the Publications
Manager, Kira Brunner. The original report, which forms the core of
Chapters 5, 8, and 9, was written at the request of and with the
support of the Commonwealth Secretariat. It was reviewed at a meet-
ing of the IPD Trade Task Force, chaired by Dani Rodrik of Harvard
and Andres Rodriguez-Clare of the Inter-American Development
Bank (both acting in a personal capacity), in New York in March
2004. There were subsequent presentations in Washington in the
spring of 2004, on the fringes of the IMF and World Bank meetings
and at a meeting held at the Institute for International Economics,
with comments presented by Supachai Panitchpakdi, the Director
General of the WTO. In May 2004 it was presented in Brussels at the
Annual Bank Conference on Development Economics—Europe,
sponsored by the World Bank; and in September 2004 it was
presented at the United Nations at the invitation of the President
of the General Assembly, and at the WTO in Geneva, and at the
Commonwealth Finance Ministers’ meeting in Saint Kitts. We wish
to acknowledge the helpful comments of the participants in those
presentations, many of which have been incorporated into this
book. We are also indebted to David Vines, Simon Evenett, Andrew
Glyn, Sarah Caro, the publisher at Oxford University Press, Anya
Schiffrin for invaluable help with many areas of the production and
xiv ACKNOWLEDGEMENTS
publicity, Dan Choate and Josh Goodman for their editiorial assis-
tance, and several anonymous reviewers who provided useful com-
ments on the manuscript.
We would also like to acknowledge the financial support of the
Ford Foundation, the Macarthur Foundation, the Mott Foundation,
the Canadian International Development Agency (CIDA), and the
Swedish International Development Cooperation Agency (SIDA)
for their financial support for the Initiative for Policy Dialogue. In
particular, the IPD Trade Task Force is supported by a grant from
CIDA.
Joseph Stiglitz and Andrew Charlton
2005
Contents
3
List of Tables xvi
List of Figures xviii
Glossary xx
1 Introduction: The Story so Far 1
2 Trade Can be Good for Development 11
3 The Need for a Development Round 41
4 What has Doha Achieved? 57
5 Founding Principles: The Basis of a Fair Agreement 67
6 Special Treatment for Developing Countries 87
7 Priorities for a Development Round 107
8 How to Open up Markets 115
9 Priorities Behind the Border 133
10 What should not be on the Agenda? 141

11 Joining the Trading System 157
12 Institutional Reforms 167
13 Trade Liberalization and the Costs of Adjustment 171
Appendix 1: Empirical review of market access issues 215

Appendix 2: Empirical review of the Singapore Issues 261
References 279
Index 297
List of Tables
3
3.1 The nine trade negotiation rounds under the GATT
and WTO 42
3.2 Early estimates of income gains from the Uruguay
Round Agreement 46
7.1 Development issues in the Doha Round 112
13.1 Export changes resulting from the replacement of
GSP with MFN tariffs 179
13.2 Welfare effects from the replacement of GSP
with MFN tariffs 180
13.3 Utilization of non-reciprocal preferences granted
by the Quad countries to LDCs, 2001 181
13.4 Tariff averages for imports under MFN and GSP, 1999 182
13.5 Effect of full tariff liberalization on high-value
agricultural imports to US 185
13.6 Importance of products liberalized under the EBA 186
13.7 Tariff revenue for selected countries, 1995 190
13.8 Summary of effect of trade liberalization on revenue 191
13.9 Effect of global trade liberalization on market prices
in Indonesia 200
13.10 Effect of global trade liberalization on poverty
in Indonesia 201
13.11 Effect of global trade liberalization on poverty
for 14 developing countries 202
13.12 Trade-related assistance provided by multilateral agencies 207
A1.1 Average protection in agriculture and food, 2005 219
A1.2 Average agricultural tariff rates 220
LIST OF TABLES xvii
A1.3 Trade specialization indices, 1965–1998 221

A1.4 Share of developing country trade with OECD, 1997 222
A1.5 Income category and food trade status 226
A1.6 Welfare and efficiency gains expected from a
40% liberalization in agriculture, 2005 227
A1.7 Change in world trade volume from agricultural liberalization 228
A1.8 Welfare gains from global removal of trade barriers, 2005 229
A1.9 Change in average world prices due to comprehensive
OECD domestic support 231
A1.10 Welfare impacts of domestic support reform 232
A1.11 Estimated average rates of protection by
region and sector, 2005 240
A1.12 Welfare effects of service sector liberalization 244
A1.13 Welfare and efficiency gains from liberalization of
agriculture, manufacturing, and services 245
A1.14 Average manufacturing tariff rates 255
A1.15 Estimated welfare gains from the Uruguay reductions
in manufactures tariffs 257
A1.16 Estimated welfare gains from manufacturing
liberalization in Doha Round 258
A1.17 Estimated welfare and efficiency gains from a
40% liberalization in agriculture and manufacturing, 2005 259
List of Figures
3
6.1 WTO members’ GDP and GDP per capita 96
11.1 Length of accession process for the first 20 countries
to join the WTO 158
11.2 Average tariff binding on agricultural products permitted
to the first 20 countries to join the WTO 161
13.1 Agricultural import protection in the US 183
13.2 Average tariff revenue, 1995 189
13.3 Wage growth by country groups, 1980s–1990s 196
13.4 Changes in unskilled wages resulting from three
liberalization alternatives within the Doha Round 197
13.5 Liberalization and inequality 198
13.6 Expenditure on social security and welfare 204
13.7 Commodity structure of exports, Tanzania and Malaysia 211
A1.1 Agricultural producer support, 1986–1988
and 1999–2001 218
A1.2 Foreign direct investment in services 238
A1.3 Full and partial market access commitments under GATS 239
A1.4 Service sector openness by region: financial services
and telecommunications 241
A1.5 Share of world exports, manufacturing and
agriculture, 1965–1995 253
A1.6 Share of developing countries’ exports, manufacturing
and agriculture, 1965–2005 253
A1.7 Developing countries’ share of world trade, 1970–2000 254
A1.8 Selected developed country imports from all developing
countries, 1980–1995 254
LIST OF FIGURES xix
A1.9 Average MFN tariff on manufactures, 1995 and 2005 256

A1.10 Share of post-Uruguay global liberalization gains accruing
to developing countries 259
A2.1 Liberalization of investment regimes, 1991–2001 264
A2.2 Bilateral investment treaties, 1960s–1990s 265
A2.3 Volume of imports affected by cartels, 1981–2000 270
Glossary
3
African, Caribbean, and Pacific (ACP) countries Group of African, Caribbean,
and Pacific countries which have received special treatment from the European
Union through a series of agreements, including the Lomé Convention and the
Cotonou Agreement.
Agreement on Agriculture WTO agreement which focuses on improving

market access and reducing trade-distorting domestic support payments
and export subsidies in agriculture.
anti-dumping duties Specific import duties imposed by importing countries

on goods which are dumped by foreign exporters and cause injury to producers
of competing products.
anti-globalization A political stance of opposition to the perceived negative

aspects of globalization.
Appellate Body The WTO’s judicial body that hears appeals to the findings
of dispute settlement panels.
Cairns Group A group of countries which lobby together for agricultural

liberalization, including Argentina, Australia, Bolivia, Brazil, Canada, Chile,
Colombia, Costa Rica, Fiji, Guatemala, Indonesia, Malaysia, New Zealand,
Paraguay, the Philippines, South Africa, Thailand, and Uruguay.
competition policy Policies designed to protect and stimulate competition

in markets by outlawing anti-competitive business practices such as cartels,
market sharing, or price fixing, the body of laws of a state which encourage
competition by restricting practices which remove competition from the
market, such as monopoly and cartels.
compulsory license Authorization for a government or company to make

and sell a product (such as a life-saving drug) without the permission of the
patent holder on the grounds of public interest.
Cotonou Agreement A treaty signed in Cotonou, Benin, in June 2000

which sets out the relationship between the European Union and the ACP
GLOSSARY xxi
countries on foreign aid, trade, investment, human rights, and governance;

Replaces the Lomé Convention.
countervailing duty A means to restrict international trade in cases

where imports are subsidized by a foreign country and harm domestic
producers.
development box Measures proposed to give developing countries special

flexibility within WTO rules for the purpose of ensuring food security,
protecting farmer livelihoods, and reducing poverty.
Dispute Settlement Body The General Council of the WTO, composed

of representatives of all member countries, convened to administer rules
and procedures established in various agreements. It has the authority to
establish panels, oversee implementation of rulings and recommendations,
and authorize suspension of concessions or other obligations under various
agreements.
Doha Declaration Statement made at the fourth WTO ministerial conference

in Doha, Qatar, launching the Doha Round.
dumping The export of goods at a price less than their normal value, generally
at less than in the domestic market or third-country markets, or at less than
production cost.
enabling clause The 1979 decision of the GATT to give developing countries
‘differential and more favorable treatment, reciprocity and fuller participa-
tion’. One of the so-called framework agreements, it enables WTO members
to accord such treatment to developing countries without giving it to other
contracting parties.
Everything but Arms (EBA) The name given by the EU to the package it
offered to the least developed countries in 2001, which is expected to eliminate
quotas and tariffs on all of their exports—except arms.
externality A side-effect or consequence (of an industrial or commercial

activity) which affects other parties without this being reflected in the cost
of the goods or services involved; a social cost or benefit.
G33 A group actually consisting of 42 developing countries of the WTO.

They are: Antigua and Barbuda, Barbados, Belize, Benin, Botswana, China,
Republic of the Congo, Côte d’Ivoire, Cuba, Dominican Republic, Grenada,
Guyana, Haiti, Honduras, India, Indonesia, Jamaica, Kenya, Korea, Mauritius,
Mongolia, Montserrat, Mozambique, Nicaragua, Nigeria, Pakistan, South
Panama, Peru, the Philippines, Saint Kitts, Saint Lucia, Saint Vincent and
xxii GLOSSARY
the Grenadines, Senegal, Sri Lanka, Suriname, Tanzania, Trinidad and Tobago,
Turkey, Uganda, Venezuela, Zambia, and Zimbabwe.
General Agreement on Tariffs and Trade (GATT) An organization established

in 1947 to agree on common rules for tariffs and to reduce trade restrictions
through a series of negotiating rounds. The Uruguay Round, completed in
1994, created the World Trade Organization, which superseded the GATT
in 1995.
General Agreement on Trade in Services (GATS) WTO agreement concluded

at the end of the Uruguay Round. It provides a legal framework for trade
in services, and the negotiated, progressive liberalization of regulations
that impede this. It covers areas such as transport, investment, education,
communications, financial services, energy and water services, and the
movement of persons.
Generalized System of Preferences (GSP) A program to grant trade advantages

(such as reduced tariff levels) to particular developing countries.
government procurement Purchase of goods and services by governments

and state-owned enterprises.
green box Income support and subsidies that are expected to cause little or
no trade distortion. The subsidies have to be funded by governments but
must not involve price support. Environmental protection subsidies are
included. No limits or reductions are required for such income support or
subsidies.
Green Room Closed meetings during which developed countries negotiated

with selected countries as part of non-transparent bargaining tactics during
the GATT and WTO proceedings.
import quota A form of protectionism used to restrict the import of goods

by limiting the legal quantity of imports.
import substitution A trade and economic policy based on the premise that a
developing country should attempt to substitute locally produced substitutes
for products which it imports (mostly finished goods). This usually involves
government subsidies and high tariff barriers to protect local industries
and hence import substitution policies are not favored by advocates of
absolute free trade. In addition import substitution typically advocates an
overvalued currency, to allow easier purchase of foreign goods, and capital
controls.
GLOSSARY xxiii
infant industry protection Protection of a newly established domestic

industry.
Jubilee 2000 An international coalition which called for cancellation of

unpayable third world debt by the year 2000.
market access The extent to which a country permits imports. A variety of

tariff and non-tariff trade barriers can be used to limit the entry of products
from other countries.
market failure A case in which a market fails efficiently to provide or

allocate goods and services, therefore requiring state intervention.
Mercosur A trading zone consisting of Brazil, Argentina, Uruguay, and

Paraguay, founded in 1995. Its purpose is to promote free trade and movement
of goods, peoples, skills, and money between these countries.
Millennium Development Goals (MDGs) Goals which governments

committed themselves at the UN General Assembles in 2000 to achieving
by 2015: namely, eradicating extreme poverty and hunger; achieving
universal primary education; promoting gender equality and empowering
women; reducing child mortality; improving maternal health; combating
HIV/AIDS; malaria, and other diseases; ensuring environmental sustain-
ability; and developing a global partnership for development.
mode of supply WTO term to identify how a service is provided by a

supplier to a buyer.
most-favored-nation (MFN) treatment A country extending to another

country the lowest tariff rates it applies to any other country. All WTO
contracting parties undertake to apply such treatment to one another
under Art. I of the GATT. When a country agrees to cut tariffs on a particular
product imported from one country, the tariff reduction automatically
applies to imports of that product from any other country eligible for most-
favored-nation treatment.
national treatment Treating foreign producers and sellers the same as

domestic firms.
necessity test Procedure to determine whether a policy restricting trade is

necessary to achieve its intended objective.
non-tariff barriers (NTBs) A catch-all phrase describing barriers to interna-

tional trade other than tariffs.
xxiv GLOSSARY
Organization for Economic Co-operation and Development (OECD) Group

of industrial countries that ‘provides governments a setting in which to
discuss, develop and perfect economic and social policy’.
parallel imports Products made and marketed by the patent owner (or trade-
mark or copyright owner) in one country and imported into another country
without the approval of the patent owner.
Pareto efficiency The criterion which stipulates that for change in an

economy to be viewed as socially beneficial, it should make no one worse
off while making at least one person better off.
patent A grant from a government to a firm conferring the exclusive

privilege of making or selling some new invention.
Poverty Reduction Strategy Paper (PRSP) A document describing a country’s

macroeconomic, structural and social policies and programmes to promote
growth and reduce poverty, as well as associated external financing needs.
Initiated by the boards of the World Bank and International Monetary
Fund (IMF), PRSPs are expected to be prepared by governments through
a participatory process involving civil society and development partners,
including the World Bank and IMF, and are required for countries seeking
to obtain concessional lending and debt relief under the enhanced Heavily
Indebted Poor Countries (HIPC) initiative.
predatory pricing Action by a firm to lower prices so much that rival firms
are driven out of business, after which the firm raises prices to exploit its
resulting monopoly power.
production subsidy A payment perhaps implicit, by government, to producers

encouraging and assisting their activities and allowing them to produce at
lower cost or to sell at a price lower than the market price.
protocol of accession Legal document recording the conditions and obliga-

tions under which a country accedes to an international agreement or
organization.
Quad countries Canada, the EU, Japan, and the US.
quota Measure restricting the quantity of a good imported or exported.

Quantitative restrictions include quotas, non-automatic licensing, mixing
regulations, voluntary export restraints, and prohibitions or embargos.
Rules of Origin Criteria for establishing the country of origin of a product.

Often based on whether production (processing) leads to a change in tariff
GLOSSARY xxv
heading (classification) or in the level of value added in the country where

the good was last processed.
safeguard action or measure Emergency protection to safeguard domestic

producers of a specific good from an unforeseen surge in imports.
sanitary and phytosanitary measures Border control measures necessary to

protect human, animal, or plant life or health.
second-best argument for protection An argument for protection to partially

correct an existing distortion in the economy when the first-best policy for
that purpose is not available.
Singapore Issues The topics discussed by four working groups set up

during the WTO Ministerial Conference of 1996 in Singapore, namely
investment protection, competition policy, transparency in government
procurement, and trade facilitation. Disagreements, largely between
largely developed and developing economies, prevented a resolution
of these issues, despite repeated attempts to revisit them, notably during
the 2003 Ministerial Conference in Cancún, Mexico, where no progress
was made.
single undertaking Provision that requires countries to accept all the agree-
ments reached during the Uruguay Round negotiations as a single package,
rather than on a case-by-case basis.
special and differential treatment The principle in the WTO that developing
countries be accorded special privileges, either exempting them from
some WTO rules or granting them preferential treatment in the application
of WTO rules.
tariff A government-imposed tax on imports.
tariff binding Commitment not to increase a rate of duty beyond an agreed

level. Once a rate of duty is bound, it may not be raised without compensating
the affected parties.
tariff equivalent The level of tariff that would be the same, in terms of its
effect, as a given non-tariff barrier.
tariff escalation An increase in tariffs as a good becomes more processed,

with lower tariffs on raw materials and less processed goods than on more
processed versions of the same or derivative goods: for example, low duties
on fresh tomatoes, higher duties on canned tomatoes, and higher yet on
tomato ketchup.
xxvi GLOSSARY
tariffication Conversion of non-tariff barriers to their tariff equivalents.
tariff peak A single, particularly high, tariff on a good.
tariff rate quotas (TRQs) The quantitative level of imports of agricultural

products (quotas) above which higher tariffs are applied.
terms of trade Ratio of the price of a country’s export commodities to the

price of its import commodities. An improvement in a nation’s terms of
trade is good for that country in the sense that it has to pay less for the
products it imports, that is, it has to give up fewer exports for the imports
it receives.
trade diversion Trade displacement, as a result of trade policies that dis-

criminate among trading partners, of more efficient (lower-cost) sources by
less efficient (higher-cost) sources. Can arise when some preferred suppliers
are freed from barriers but others are not.
trade liberalization Reduction of tariffs and removal or relaxation of non-

tariff barriers.
Trade-Related Aspects of Intellectual Property Rights (TRIPS) Agreement

WTO agreement which sets down minimum standards for most forms
of intellectual property regulation within all member countries of
the WTO.
Trade-Related Investment Measures (TRIMS) Agreement WTO agreement

aimed at eliminating the trade-distorting effects of investment measures
taken by members. It does not introduce any new obligations, but merely
prohibits TRIMS considered inconsistent with the provisions of GATT
1994 for both agricultural and industrial goods.
Uruguay Round The last round under the GATT, which began in Uruguay
in 1986 and was completed in 1994 after nearly eight years of negotiations.
It created the World Trade Organization.
value added Additional value created at a particular stage of production, i.e.

the value of output minus the value of all inputs used in production.
voluntary export restraint An agreement between importing and exporting

countries in which the exporting country restrains exports of a certain product
to an agreed maximum within a certain period.
World Trade Organization An International organization which administers

multilateral agreements defining the rules of international trade between
GLOSSARY xxvii
its member states. The WTO replaced the General Agreement on Tariffs and
Trade; its primary mission is to reduce international trade barriers.
WTO-plus Trade agreements that go beyond what the WTO multilateral

trade regime requires. Regional trade agreements sometimes contain
WTO-plus elements.
1
Introduction: The Story so Far
3
In November of 2001, trade ministers from 140 nations gathered in
Doha, Qatar, to seek to give the World Trade Organization (WTO) a
historic new mandate. The WTO’s previous ministerial meeting in
Seattle, USA, in 1999 had ended in failure and brought protests and
violence to the city. Now they were meeting in the midst of a global
recession, just two months after the shocking attacks on the United
States of 11 September 2001. On the evening of 14 November, after
several days of negotiations, and more than 18 hours after their
original deadline, the ministers announced that they had reached a
landmark agreement to launch a new round of trade talks. The
agreement—the Doha Declaration—outlined a framework for a
wide-ranging new round of multilateral negotiations. The top US
trade negotiator, Robert Zoellick, was jubilant. ‘We have sent a pow-
erful signal to the world,’ he said, adding that a new trade round
would deliver ‘growth, development and prosperity’.1 Zoellick’s
claim that a new round of trade liberalization would deliver prosper-
ity to the world was plausible, but he was perhaps too optimistic
about the ensuing negotiations.
The purpose of this book is to describe how trade policies can be
designed in developed and developing countries with a view to integ-
rating developing countries into the world trading system and to
help them to benefit from their participation. This book starts from
the presumption that trade can be a positive force for development.
1
Quoted in ‘Seeds Sown for Future Growth’, The Economist, 15 Nov. 2001.
2 FAIR TRADE FOR ALL
In the right circumstances, policies which reduce tariffs and other

barriers to the movement of goods and services can facilitate trade
between nations and deliver welfare gains. However, we also point
out that while increased trading opportunities are good for develop-
ing countries, liberalization needs to be managed carefully—the task
is much more complex than the simple prescriptions of the
Washington Consensus, which blithely exhorts developing coun-
tries to liberalize their markets rapidly and indiscriminately.2
In the run-up to the Doha meeting, the expectations of the devel-
oping country members of the WTO had been tempered by negative
experiences from previous rounds of trade negotiations. Many devel-
oping countries had feared that the large industrialized countries
would use their superior bargaining power to force through agree-
ments which would disadvantage the developing countries. These
fears seemed to be realized in the Uruguay Round. After the round
was completed and an agreement had been signed, developing
countries felt that they had not been fully aware of the cost of the
obligations they had agreed to, and they felt that the liberalization
in developed countries had fallen short of their expectations.
Developing countries gained less than they had hoped on such issues
as the speed with which textile protection would be reduced and the
extent of tariff and subsidy reduction on agricultural goods in devel-
oped countries. The developed countries walked away from Uruguay
with a large share of the gains, and many developed countries were
predicted to be left worse off as a result of the round. After the
Uruguay Round, many developing countries were reluctant to
embark upon a new round of trade negotiations which might lead to
another imbalanced agreement.
However, at the turn of the millennium there was renewed
optimism. A new global consensus seemed to be developing to confront
the economic challenges faced by the poorest nations which gained
prominence in international affairs through a series of new initiatives.
In trade, the Cotonou Agreement, led by the European Union, and
2
The Washington Consensus is a set of policies believed by some economists to be the formula for pro-
moting economic growth in developing countries. These policies include privatization, fiscal discipline, trade
liberalization, and deregulation. In the 1990s these policies were vigorously advocated by several powerful
economic institutions located in Washington, including the International Monetary Fund, the World Bank,
and the US Treasury.
INTRODUCTION: THE STORY SO FAR 3
the US African Growth and Opportunity Act (AGOA) granted exporters

from the poorest countries in the world free access to the richest
markets in the world. At the same time, grassroots movements—
including the worldwide Jubilee 2000 campaign devoted to debt
cancellation for the poorest countries and the World Social Forum—
gained unprecedented publicity and momentum for their causes. In
addition, world leaders signed landmark treaties which placed the
plight of poor countries at the heart of the global agenda. At the
United Nations Millennium Summit in New York in September 2000,
world leaders adopted the Millennium Development Goals (MDGs);3
at the International Conference on Financing for Development, held in
Monterrey, Mexico in March 2002, the advanced industrial countries
committed themselves to helping provide the finance for develop-
ment priorities of poor countries; and the Johannesburg Summit on
Sustainable Development in September 2002 established an action
plan to ensure sustainable global development.
The WTO conference at Doha reflected the new determination to
address development problems collectively in multilateral forums.
It was a hopeful milestone for developing nations and they emerged
from it with some optimism. Several of their concerns were incorpo-
rated into the agreement made at Doha, which explicitly focused on
the promotion of economic development and the alleviation of
poverty in poor countries. This ‘Doha Round’—the ninth of a series
of such negotiations, and the first since the formalization of trade
negotiations under the WTO4—came to be commonly referred to as
the ‘Development Round’.
Unfortunately, in the years since it was launched, the Doha Round
has not delivered on its development mandate in several important
respects. First, there has been little progress on the issues of interest
to developing countries. In particular, developing countries are
interested in agreements to reduce tariffs on the goods that they can
3
The Millennium Development Goals are the United Nations targets for reducing poverty, hunger, dis-
ease, illiteracy, environmental degradation, and discrimination against women by 2015. The targets address
extreme poverty in its many dimensions—income poverty, hunger, disease, lack of adequate shelter, and
exclusion—while promoting gender equality, education, and environmental sustainability. They are also
basic human rights—the rights of each person on the planet to health, education, shelter, and security.
4
The first round, held in Geneva in 1947, resulted in the General Agreement on Tariffs and Trade (GATT),
which was formally replaced by the WTO in 1995.
export competitively. These are mainly labor-intensive goods, i.e.

goods that are produced cheaply in countries with low wage rates
and abundant unskilled labor.
A second problem with the so-called ‘Development Round’ is that
the new issues that were initially put on the agenda primarily reflect
the interests of the advanced industrial countries and have been
strongly opposed by many developing countries. The most promin-
ent new issues in the Doha Round emerged from the decision by
WTO member countries at the 1996 Singapore Ministerial Conference
to establish three new working groups: on trade and investment, on
competition policy, and on transparency in government procurem-
ent. They also instructed the WTO Goods Council to consider ways
of simplifying trade procedures, an issue sometimes known as ‘trade
facilitation’. Because these four issues were introduced to the agenda
at the Singapore ministerial meeting, they are often called the
‘Singapore Issues’. These issues have been opposed by developing
countries, who are skeptical of new multilateral obligations which
could restrict existing developmental domestic policy options and
which might require large implementation costs.
Less than two years after the Doha Declaration, it had become
clear that the Round was seriously off track. In September 2003, the
WTO convened another ministerial meeting in Cancún, Mexico,
with the special task to ‘take stock of progress in the [Doha
Development Agenda] negotiations, provide any necessary political
guidance and take decisions as necessary’.5 After four days the meet-
ing ended abruptly without agreement on any of the main issues.
The apparently irreconcilable conflict between developed and devel-
oping countries which produced failure at Cancún led to calls for a
reassessment of the direction of global trade negotiations. Many of
the participants in the Cancún meeting felt that Europe and the
United States had reneged on the promises that had been made at
Doha, emblematized by the lack of progress on agriculture.
There were mutual recriminations about who was to blame for the
failure at Cancún. There was even disagreement about who would
5
This is spelt out in para. 45 of the declaration that ministers issued at the previous ministerial confer-
ence in Doha in November 2001.
suffer the most. The United States and Europe were quick to assert
that it was the developing countries who were the ultimate losers.6
But many developing countries had taken the view that no agree-
ment was better than a bad agreement, and that the Doha Round was
rushing headlong (if any trade agreement can be described as ‘rush-
ing’) into one which, rather than redressing the imbalances of the
past, would actually make them worse. Though some progress had
been made in addressing the concerns about the manner in which
the negotiations were conducted, the failure to address these con-
cerns fully7 generated the further worry that the developing coun-
tries would, somehow, be strong-armed in the end into an agreement
that was disadvantageous to them. There were also threats, espe-
cially by the United States, that it would effectively abandon the
multilateral approach, taking up a bilateral approach. It differenti-
ated between the ‘can do’ countries and others, and suggested that
the ‘can do’ countries would benefit from a series of bilateral agree-
ments. The smaller developing countries recognized that in these
bilateral discussions their bargaining position was even weaker than
it was in the multilateral setting. Several of the bilateral trade agree-
ments made since Cancún have shown that these worries were jus-
tified. On the other hand, the United States has not succeeded in
establishing a bilateral trade agreement with any major developing
country.
This book takes a step back from these disputes. It attempts to
support progress in the current round by asking what a true
Development Round of trade negotiations would look like, one that
reflects the interests and concerns of the developing countries and is
designed to promote their development. What would an agreement
based on principles of economic analysis and social justice—not on
economic power and special interests—look like? Our analysis
concludes that the agenda would look markedly different from
that which has been at the center of discussions for the past
6
See the op-eds in The Financial Times and the New York Times e.g. Robert B. Zoellick, ‘America Will Not
Wait for the Won’t-Do Countries’, The Financial Times, 22, Sept. 2003, 23.
7
Most notable in this regard was the request by a number of developing countries that the Cancún draft
be prepared on the basis of views and inputs at open-ended consultations, and where there was no con-
sensus, to indicate clearly the differing positions or views. The proposal was rejected by a coalition of devel-
oped countries.
two years, and that the fears of the developing countries that the
Doha Round of trade negotiations would disadvantage them (were
the demands of the developed countries acceded to) were in fact
justified.
In Chapter 2 we describe the conceptual foundations for the policy
proposals in this book, starting with the theoretical proposition that
trade liberalization is, in general, welfare-enhancing. We then show
that for some countries, particularly the least developed countries,
the assumptions on which this proposition is based are not entirely
applicable. The problems of poverty, inequality, and incomplete risk
and capital markets lead the experience of these countries to diverge
from the predictions of the simple neo-classical models. These prob-
lems cause the experience of liberalization to vary across countries
depending on their individual characteristics. There is a difference
between trade openness (the state of having low barriers to imports)
and trade liberalization (the process of reducing those barriers). Trade
liberalization is supposed to deliver gains as resources are transferred
from protected sectors, in which a country does not have comparative
advantage, to those sectors where it is more efficient and where it can
export more successfully. But in developing countries the lack of
resources (labor and other production inputs) available to new indus-
tries is not usually the constraint which prevents the growth of new
export sectors.8 Developing countries have vast reserves of resources,
particularly labor, which are already unemployed or underemployed.
Thus trade liberalization is not required to ‘free up’ these resources
for use in new industries. Unless complementary policies are used to
ease the other constraints which prevent the advancement of export
sectors, liberalization, by removing the protection given to domestic
industries, may just leave the workers and other resources used in for-
merly protected industries idle in the short run.
Chapter 3 addresses the need for a Development Round. It exam-
ines some elements of the experience of developing countries in pre-
vious trade negotiations and briefly reviews some of the potential
gains available from further liberalization. Chapter 4 is a brief review
8
Other constraints on the ability of developing countries to develop successful export industries might
include technological backwardness, workforces too small to generate economies of scale, high trade and
transport costs, poor infrastructure, weak government institutions, and a lack of skilled labor.
of the Doha Round so far, and the extent to which it has lived up to
the expectations of developing countries. It makes clear that there is
a huge discrepancy between the Development Round trade agenda,
both as it was formulated at Doha and as it has evolved since, and a
true Development Round agenda that would do much more to integ-
rate the developing countries into the world trading system and to
remove the barriers which curtail the benefits they receive from
their participation. Such an agenda would promote growth in devel-
oping countries and work to reduce the huge disparity that separates
them from the more advanced industrial countries.
Chapter 5 outlines our proposals for the principles that should be
adopted in a Development Round of trade negotiations. The primary
principle of the Doha Round must be to ensure that the agreements
promote development in poor countries. To make this principle
operational the WTO needs to foster a culture of robust economic
analysis to identify pro-development proposals and promote them to
the top of the agenda. In practice this means establishing a source of
impartial and publicly available analysis of the effects of different
initiatives on different countries and groups within countries. This
should be a core responsibility of an expanded WTO Secretariat. On
the basis of this analysis, any agreement that differentially hurts
developing countries or provides disproportionate benefits to devel-
oped countries should be presumptively viewed as unfair and as
being against the spirit of the Development Round.
The agreements must enshrine both de jure and de facto fairness.
This means ensuring that developing countries are not prevented
from unlocking the benefits of free trade because of a lack of institu-
tional capacity. In this regard, developing countries will require
special assistance to enable them to participate equally in the WTO.
The principle of fairness should also be sensitive to countries’
initial conditions. Chapter 6 discusses special treatment for develop-
ing countries which is needed to recognize that adjustment to new
trading rules involves particularly high costs for those developing
countries whose institutions are weakest and whose populations are
most vulnerable. Prescriptive multilateral agreements must not be
allowed to run roughshod over national strategies to deal with
idiosyncratic development problems.
Chapters 7–10 present pro-development priorities that should

form the core of the Doha Round agreements. Much of the recent dis-
cussion has focused on agriculture, but there is much more to a true
Development Round. Primary attention should be given to market
access for goods produced by developing countries. There is an
urgent need to reduce protection on labor-intensive manufactures
(textiles and food processing) and unskilled services (maritime and
construction services). Priority should also be given to the develop-
ment of schemes to increase labor mobility—particularly the facil-
itation of temporary migration for unskilled workers. As tariff
barriers have come down, developed countries have increasingly
resorted to non-tariff barriers; these need to be circumscribed. The
proposals in these chapters are motivated by empirical analysis of
the gains and costs of liberalization. For ease of exposition the analy-
sis of this evidence is presented separately in Appendices 1 and 2.
Chapter 11 considers the terms on which non-members are able to
accede to the WTO.
Chapter 12 takes a brief look at some institutional reforms that
might facilitate a more transparent and democratic negotiating
process, and ones which might more likely result in agreements that
are both fair and in the general interests of the world. A fair agree-
ment is unlikely to be produced through an unfair process. In particu-
lar, greater transparency and openness are required to create a more
inclusive bargaining process.
Chapter 13 considers the potentially costly process of adjustment
to the kind of new trading regime envisioned in this book. In one
sense, adjustment costs can be thought of as the price to be paid for
the benefits of multilateral trade liberalization. It is these adjust-
ment costs together with the trade benefits that determine the net
effect of trade reform for each country. The Doha Round has placed
renewed emphasis on the importance of sharing the benefits of trade
reform fairly among developed and developing countries. However,
there has been less attention to the distribution of adjustment costs
among countries. The fact that implementation and adjustment
costs are likely to be larger in developing countries, unemployment
rates are likely to be higher, safety nets weaker, and risk markets poorer,
are all facts that have to be taken into account in trade negotiations.
For some of the smallest and poorest states, the adjustment costs
of trade liberalization may significantly outweigh the benefits
available.
If the Development Round is to bring widespread benefits to people
living in developing countries—and if there is to be widespread sup-
port for the continuing agenda for trade reform and liberalization—
the developed world must make a stronger commitment than it has
in the past to giving assistance to the developing world. Assistance is
required not only to help bear the often large costs associated with
trade reform, but also to enable developing countries to avail them-
selves of the new opportunities provided by a more integrated global
economy.
2
Trade Can Be Good for
Development
3
International trade can have a significant positive effect on economic
growth and development.
In the eighteenth century, technological breakthroughs put
Britain on the path to becoming the first truly ‘modern’ economy.
Between 1870 and 1950 Britain’s population nearly tripled. Towns
like Birmingham, Liverpool, and Manchester grew into huge cities,
average incomes grew more than twofold, and the share of farming
fell from just under a half to less than a fifth of total production.
There were many social, political, and geographical factors that
caused the Industrial Revolution, but Britain’s trade with her neigh-
bors and colonies played a decisive role in fueling the new industrial
activity and spreading prosperity to other countries. Before long
British cities became the workshops of the world, importing vast
quantities of food and raw materials, and exporting manufactured
goods to America, Asia, and Africa.
Meiji Japan’s rapid industrialization in the early twentieth century
was also the result of a combination of domestic and international
factors. The Meiji rulers established stable political institutions and
they were quick to adopt the Western technology they had seen dur-
ing the Iwakura missions to Europe and the United States in the
1870s. They established a new education system for all young peo-
ple, sent students to the United States and Europe, and emphasized
modern science, mathematics, technology, and foreign languages.
The government built railroads, improved the road network, and

pursued land and financial sector reforms. The availability of trading
opportunities was also vitally important. It is hard to imagine the
Meiji industrialization occurring if Japan had not been able to import
vast quantities of machinery, transport equipment, and other capital
goods from the West in exchange for exports of cheap cloth, toys, and
other labor-intensive consumer products. And this trade would have
been impossible if it were not for the steady flow of food and cheap
raw materials arriving in Japan from its colonies in Taiwan and
Korea.
Similarly, international trade played a major role in the industrial
development of North America and Australia in the nineteenth cen-
tury, and of the East Asian ‘Tiger’ economies, India, and China at
various points in the second half of the twentieth century. These
examples, together with the many instances where growth did not
occur, show that trade was necessary for sustained industrial devel-
opment, but it was not sufficient. Trade liberalization created oppor-
tunities for economic development, but other factors determined
the extent to which those opportunities were realized.
This chapter lays the foundations for the policies that we propose
later in the book. The notion that trade—free trade, unencumbered
by government restrictions—is welfare-enhancing is one of the most
fundamental doctrines in modern economics, dating back at least to
Adam Smith (1776) and David Ricardo (1816). But the subject has
always been marked by controversy because the issue facing most
countries is not a binary choice of autarky (no trade) or free trade, but
rather a choice among a spectrum of trade regimes with varying
degrees of liberalization.
Almost every country today imposes some trade restrictions and
taxes. Since World War II, the world has been moving gradually
towards reducing tariffs and restrictions on trade. Some of the devel-
oped countries that have been the most ardent advocates of trade lib-
eralization have been somewhat duplicitous in their advocacy. They
have negotiated the reduction of tariffs and the elimination of sub-
sidies for the goods in which they have a comparative advantage, but
are more reluctant to open up their own markets and to eliminate
their own subsidies in other areas where the developing countries
TRADE CAN BE GOOD FOR DEVELOPMENT 13
have an advantage. As a result we now have an international trade

regime which, in many ways, is disadvantageous to the developing
countries. In a world in which many see global poverty—by some
estimates there are more than 2 billion people living on less than a
dollar a day—as the world’s most pressing problem, this is especially
disturbing. It seems obvious that if the developed countries truly
wanted to promote development in the Doha Round they should
reduce their tariffs and subsidies on the goods of interest to the devel-
oping countries.
But many of the developed countries’ negotiators have turned this
argument on its head. They suggest that the reduction of one’s own
tariffs is beneficial, and hence the developing countries would be help-
ing themselves by liberalizing in the WTO, irrespective of the actions
of the developed countries. On this basis they argue that the develop-
ing countries should accept almost any offer that is put on the table.
If matters were so easy a pro-development trade agenda would be
trivial—the developing countries should simply unilaterally open
up their markets, and the faster they do so, the better. In this book we
argue that matters are not so easy and that a pro-development agenda
is more complex.
This chapter provides the conceptual framework for the policies
that we propose for the Doha Round. In the first section we take a
quick look at the interpretations and misinterpretations of the con-
trasting trade policies and growth experiences of Latin America and
East Asia. In the second section we examine the theory behind the
claims that trade is good for welfare and good for growth. In the third,
we turn to the difficulties confronting the empirical studies and
explain why they remain such a subject of controversy. In the final
two sections we discuss the policy implications of the theoretical
analysis and empirical evidence, and we consider the way forward. In
each section, we develop our theme that trade liberalization can pro-
mote development, but that the results of different trade policies have
varied across countries; and the evidence suggests that the benefits of
liberalization depend on a host of factors. Thus the implementation of
trade liberalization needs to be sensitive to national circumstances.
The sequencing of liberalization is important, and there is much that
can be done in conjunction with trade liberalization (by developing
and developed countries alike) to ensure that developing countries are

provided with meaningful new trading opportunities, and that they
are able to benefit from them.
The lessons of East Asia and Latin America
The post-World War II world has seen several grand experiments in

trade policy. The import substitution policies of Latin America and
the export-oriented strategies of several East Asian countries are of
particular interest. In both cases several countries undertook similar
policies at similar times and saw broadly similar results. Economists
have attempted to draw trade policy lessons from these experiences,
but the conclusions have been contentious because each country
pursued a multifaceted economic policy agenda from which it is dif-
ficult to isolate the contribution of trade policy alone to their eco-
nomic success or failure.
The success of East Asia

The success of East Asia begins with Japan, which within a few
decades after World War II had raised itself to the second largest
economy in the world. It experienced sustained growth rates beyond
those previously seen anywhere in the world. Japan’s success was
followed first by South Korea, Taiwan, Hong Kong, and Singapore;
and then by Thailand, Indonesia, and Malaysia, and finally by China.
East Asia’s growth over more than three decades was remarkable.
In particular Japan and the other countries in East Asia refuted two
of the classic propositions of development. First, they showed that
inequality was not necessary to growth, whereas previously it had
been widely believed that only through inequality could the requi-
site high savings rates be generated (Lewis 1955). Second, these
countries proved that the initial stages of development did not have
to be associated with an increase in inequality (contra Kuznets
1955). Instead the new prosperity was widely shared among its
population and millions were lifted out of poverty. For example in
Malaysia and Thailand, the incidence of poverty declined from
almost 50 per cent in the 1960s to less than 20 per cent by the end of
the century.
There were important differences among the polices pursued by
the East Asian countries. Korea and Japan’s industrial model focused
on large domestic corporate conglomerates and actively restricted
flows of foreign direct investment (FDI), which made up less than 5
per cent of GDP in the period 1987–92. By contrast Singapore and
Malaysia both developed policies to attract large foreign multina-
tional corporations and encourage clusters of activity to develop
around them. FDI in these countries reached more than 30 per cent
of GDP by 1992. However, at a deeper level, the Asian countries
shared much in common. They had high rates of investment in phys-
ical and human capital, rapid growth of agricultural productivity,
and declining fertility.
But it is the role of the state in Asia’s growth miracle which has
created the most controversy. Adherents of the orthodox view
believe that East Asia’s free market philosophy was the main source
of its success. They stress the prevailing stability-oriented macro-
economic policies, including responsible government monetary and
fiscal policy, low inflation, and the maintenance of an appropriate
real exchange rate. They also stress the reliable legal framework,
which promoted stability, investment, and competition.
These factors were certainly conducive to growth, but they are far
from the whole story. In other respects the Asian countries’ eco-
nomic policies certainly did not fit the orthodox framework. Their
economic model included a strong role for the public sector. They
clearly did not believe in free and unfettered markets. At the core of
success had to be well-functioning firms and markets, but govern-
ment played a critical role. Governments acted as catalysts which
helped markets by providing the requisite physical and institutional
infrastructure, by remedying market failures, and by promoting sav-
ings and technology. Noland and Pack (2003) trace Japan’s industrial
policies back to the Meiji Restoration of the mid-nineteenth century,
and the state-led development under the slogans ‘shokusan-kogyo’
(industrialization) and ‘fukoku-kyohei’ (a wealthy nation and a

strong army). They point out that the economic terms of the treaties
forced on Japan by the Western powers encouraged the development
of state intervention. The rather onerous treaties required Japan to
reduce its tariffs, and consequently encouraged Japanese policy-
makers to formulate alternative ways of providing support to their
domestic industries, including subsidized credit from state banks.
These policy tools survived into the last decades of the twentieth
century.
In the modern context, there are many examples of government
selectively intervening in particular industries, although the con-
sequences of these interventions is controversial. For example, the
Japanese government initially cultivated heavy industries in the
post-war period. The steel, aluminum, car, and shipbuilding indus-
tries all received support after the war and in subsequent decades
more advanced industries, including electronics and semi-conductors,
were supported as the government expanded credit to large firms for
the purpose of fostering investment.
In many countries trade policy in particular did not follow the
orthodox free trade prescriptions. The governments of many Asian
countries pursued a two-track policy of protection for industries not
ready to compete internationally and promotion for export-ready
industries. For example, governments intervened in many industries
by providing credit through banks supported in one way or the other by
government, restricting competition from imports, constraining new
domestic competitors, and developing export marketing institutions.
These elements of the East Asian policies certainly did not fit the
orthodox ‘laissez-faire’ economic model. But there is controversy
about the role of these industrial and trade policies, with both those
who argue for and against government interventions claiming that
East Asia supports their case. Proponents of laissez-faire economic
policies contend that the industrial policies were irrelevant or even
harmful. Some economists argue that total-factor productivity
growth in the sectors which were supported by government policies
that have been the beneficiary of industrial policy has not been par-
ticularly strong. But the methodologies for calculating total-factor
productivity are notoriously weak, especially at the sectoral level. In
addition, to the extent that industrial policies in one sector led to

improved productivity in other sectors (so-called ‘spillover effects’),
the benefits of these policies would not be entirely captured by
sectoral total-factor productivity. Still other economists argue that
growth would have been even stronger had government not engaged
in these industrial policies. But this is a particularly unpersuasive
counterfactual, since no country has ever achieved faster sustained
growth than the countries of East Asia (Stiglitz 1996).
Although there is debate about the role of industry policy in Asia’s
success, there can be no doubt that the policies pursued by these coun-
tries were broader than (and in some cases clearly violated) the strict
free-market prescriptions of the Washington Consensus. Whatever
one’s beliefs about the desirability of active industry policy, there can
be no doubt that there is ongoing controversy and debate about the role
of trade policy, industry policy, and controls on capital flows (including
regulation of foreign direct investment). There can also be no doubt
that the successful cases of development over the last fifty years have
pursued inventive and idiosyncratic economic policies. To date, not
one successful developing country has pursued a purely free market
approach to development.
In this context it is inappropriate for the world trading system to
be implementing rules which circumscribe the ability of developing
countries to use both trade and industry policies to promote indus-
trialization. The current trend to force a narrow straitjacket of policy
harmonization on developing countries is simply not justified by the
available evidence. Economists have learned much about the
process of economic development, but there is still a lot that we do
not know, and in these areas developing countries should be given
the freedom to develop their own policy strategies tailored to their
own idiosyncratic circumstances.
Latin America and import substitution

In the years following the Second World War, Latin America tried a
quite different economic strategy than East Asia did. Like many
third world countries, several Latin American governments took
heart from the recent experiences of the richer countries. Many of

the countries that had fought in the Second World War had achieved
centrally planned growth in heavy industries as they mass-produced
munitions, ships, aircraft, machinery, and chemicals for the war
effort. Developing countries had also witnessed the ‘big bang’ of
Stalinist industrialization in the Soviet Union during the 1930s. The
Soviet Union experienced rapid capital accumulation and double-
digit economic growth rates while the more liberal Western capital-
ist economies floundered in the Great Depression. The apparent
industrial successes of wartime planning and the Soviet planned
economies conspired to convince many developing countries that
there was a large role for government in managing the industrializa-
tion process.
These observations were supported by development economists
who believed that the problems in developing countries were struc-
tural and required radical government intervention to overcome.
Arthur Lewis (1955) proposed that economic development required
coordination because ‘various sectors must grow in the right rela-
tionship to each other or they cannot grow at all’. He advocated a
form of managed industrialization simultaneously occurring across
many sectors to achieve ‘balanced growth’. Other economists com-
bined this idea with economies of scale and concluded that the prob-
lem of underdevelopment could only be broken by a ‘big push’ of new
investments across many sectors which would reinforce each other.
Paul Rosenstein-Rodan (1961) suggested that attempts at economic
development which were too narrowly focused on a small number of
sectors would run into the problem of inadequate demand, which
would ultimately constrain growth.
The prevailing economic wisdom was thus that economic devel-
opment required industrialization and the development of vibrant
manufacturing industries, and that industrialization would not hap-
pen on its own. At the time, developing countries’ production con-
sisted mainly of agricultural goods. Since most of the manufactured
goods consumed in these countries were imported, they came to the
conclusion that the path to success lay in encouraging domestic firms
to produce the consumer goods that had previously been acquired
from abroad. Accordingly many developing countries embarked on
‘import substitution’ policies. It was argued that they should import

only ‘essential’ capital goods. Not only would scarce foreign
exchange thereby be directed to where it had the highest social
returns, but the resulting demand for locally produced goods (because
other imports were restricted) would promote industrialization.
Moreover, only through protection could their industries compete
with the well-established firms of Europe and the United States.
In Brazil in 1951 the government of Getúlio Vargas established a
system of import licensing to give priority to imports of fuel and cap-
ital goods. They subsequently augmented this with a multiple
exchange rate system, whereby priority imports were brought in at a
favorable rate, while imports of goods that were deemed to be domes-
tically producible, were hit with higher exchange rates. Later, trade
policy was added to the mix when the Tariff Law of 1957 increased
protection for domestically produced goods. In the 1950s, 1960s, and
1970s similar import substitution policies were pursued by develop-
ing countries across the world, including Chile, India, Ghana, Peru,
Brazil, Mexico, Argentina, Ecuador, Pakistan, Indonesia, Nigeria,
Ethiopia, Zambia, and others.
Of course, the idea that these developing countries should attempt
to use trade policy to actively promote industries in which they are
uncompetitive is anathema to the simple logic of comparative
advantage that David Ricardo had elucidated more than a century
before. The reason so many countries rejected comparative advant-
age in the context of their economic development strategies lay in
the prevailing belief that the concept of comparative advantage was
insufficient because it was too static. Developing countries did not
want to rely on the primary commodity exports that were compat-
ible with their existing capabilities because they saw them as having
limited long-term growth prospects and falling terms of trade.
Instead they believed that comparative advantage could be developed
over time toward more ‘desirable’ industries with the help of active
industrial and trade policies.
Latin American countries grew rapidly in the decades of import
substitution. But then, in the early eighties, one country after
another found itself in difficulty; they defaulted on their debt, and
the continent entered ‘the lost decade’, during which growth halted
and the region’s income per person actually fell. Economic growth
rates, which had averaged 6 per cent for the region in the 1970s, fell
to almost zero in the 1980s.
The contrast between Latin America’s stagnation in the 1980s
and South-East Asia’s remarkable growth led many commentators
to draw conclusions about the relative effectiveness of their trade
policies. This stark regional contrast did not appear to be attribut-
able to resource endowments or global factors, and thus it appeared
that the difference must lie in the policies each region pursued. In
this regard, many economists believed that the major important
differences between the two groups of countries were the policies of
integration, openness, and free trade, i.e. import substitution in
Latin America versus export promotion in Asia. The neo-liberal
view was that Latin America’s problem was too much state inter-
vention in developing national industries, which caused them to be
inefficient and uncompetitive and required too much government
spending, which ultimately caused runaway inflation. The IMF and
the World Bank, in particular, championed the view that import
substitution was one of the main causes of stagnation in Latin
American countries.
Import substitution rested on the controversial belief that tempo-
rary industry support from government could promote long-term
development—often referred to as the ‘infant industry’ argument.
This analysis argues that there is some dynamic element to indus-
trial development which, when combined with a market failure, can
justify temporary government intervention. One branch of the argu-
ment suggests that firms may need to go through an initial period of
learning before they are able to compete successfully with more
established foreign firms. However, critics argued that if a firm even-
tually becomes profitable, then it should be able to finance its learn-
ing phase through private capital markets (assuming that effective
capital markets exist), and if the benefits of this learning stay wholly
within the firm then there is no case for government intervention.
Only some imperfection in the capital market justifies government
action and, even then, the best policy (if available to developing
countries) would be to improve the capital market rather than
impose trade distortions.
Another branch of infant industry theory argues that pioneering

firms bring benefits to the economy when they may invest in
providing workers with new knowledge and skills that can be appro-
priated by other firms when workers move or start their own firm.
Or alternatively, pioneering firms may generate new knowledge
which becomes a public good available to all subsequent firms.
However, the infant industry argument was criticized by Robert
Baldwin (1969), who argued that, even when market imperfections
existed, temporary industry protection might be futile. It might not
provide an incentive for firms to acquire more knowledge than they
otherwise would. Also, by subsidizing domestic production, infant
industry protection might encourage later entrants to bring their
investments forward, which could actually make the pioneer firm
worse off. Baldwin showed how some of the simplistic arguments
against free trade were theoretically flawed; but as later discussion
makes evident, some compelling arguments remained.
However, an alternative to the neo-liberal view argues that Latin
America’s failure had less to do with import substitution and more
to do with exogenous factors independent of domestic policies. The
combined effects of a global recession and the policy response of the
developed countries had a deleterious effect on the region. According
to the South Centre (1996: 42) Latin American countries simultane-
ously experienced four kinds of shocks: ‘a demand shock to developing
country exports; a consequent fall in commodity prices and a terms
of trade shock; an interest rate shock; and a capital supply shock’.
This alternative view puts the blame for the lost decade not so
much on the import substitution strategy, but on the debt policy of
Latin American countries combined with unfortunate global cir-
cumstances. These countries borrowed heavily during the seventies,
enabling them to avoid the global recession which followed upon the
oil price shock. But by the end of the 1970s the region’s foreign debt
had exploded and debt service payments reached US$33 billion per
year—nearly one-third of the region’s export earnings. Latin American
countries were left to bear the risk of interest rate fluctuations; when
the US Federal Reserve raised interest rates to unprecedented levels,
many countries were pushed over the brink. Among the evidence for
this interpretation is the fact that all of these countries, both those in
which there were relatively large problems with the import substi-
tution program and those in which there were not, went into default,
and at about the same time, shortly after the increase in American
interest rates. If the underlying problem had been the import substi-
tution strategy, then presumably the unwinding of that strategy
would have taken place differently in different countries. Yet not one
single Latin American country experienced much growth during the
1980s, regardless of their policy differences.
This alternative view suggests that it was Latin America’s open
capital markets rather than its relatively closed trade policy which
led to the lost decade. In the 1970s Latin American countries operated
the most open capital markets in the developed world—evidenced by
their high share of global FDI flows. In terms of financial liberaliza-
tion, Latin America was far more open than South East Asia, where
controls over foreign capital flows were strict. Latin America’s
reliance on foreign capital flows and foreign direct investment are
what made it particularly vulnerable to global economic shocks.
Thus, just as there are alternative interpretations of the role of
trade and industrial policy in East Asia’s success, there are alterna-
tive views on the role of trade and industrial policy in Latin America’s
failure. Certainly the import substitution policies were far from
perfect and there were some bad investments and some corruption.
But what Latin America and East Asia show is that the process of
successful liberalization is considerably more complex than the
neo-liberal Washington Consensus would suggest. The Asian coun-
tries pursued complex economic development policies which com-
bined government intervention with export promotion and controls
on the volume and quality of capital inflows. Moreover they sequenced
their liberalization and paid attention to social policy, including edu-
cation and equality, as well as investing heavily in infrastructure and
technology.
Mexico
In 1994, Mexico entered the North American Free Trade Agreement
(NAFTA), a far-reaching trade liberalization agreement with its
northern neighbors, the United States and Canada. If ever there were
an opportunity to demonstrate the value of free trade for a develop-
ing country, this was it. NAFTA gave Mexico access to the largest
economy in the world, which was right next door.
After ten years, Mexico’s experience of trade liberalization under
NAFTA has been mixed. Certainly there have been several benefits.
Trade liberalization has stimulated trade, with Mexico’s exports
growing at a rapid annual rate of around 10 per cent per year through
much of the 1990s. Foreign direct investment has also significantly
increased. And NAFTA played a critical role in Mexico’s recovery
from the Tequila Crisis of 1994–5 (Lederman, Menendez, Perry et al.
2000).
On the other hand, growth during the first decade of free trade
was slower than it had been in earlier decades (prior to 1980), mean
real wages at the end of the decade were lower, and some of the
poorest had been made worse off as subsidized American agricul-
tural products flooded the market and lowered the price received
for their domestic production. Inequality and poverty both
increased under NAFTA and by the end of the decade, Mexico was
losing to China many of the jobs that had been created since the
signing of NAFTA. Even the manufacturing sector, which had seen
significant output growth, has experienced a net loss in jobs since
NAFTA took effect.
Three lessons emerged from Mexico’s experience which are of par-
ticular relevance to our discussion in later chapters on how trade and
trade liberalization may affect development. The first is that trade
liberalization by itself clearly does not ensure growth, and its
impacts may well be swamped by other factors. Mexico suffered
from low levels of innovation—low research and development
expenditures and low levels of patenting activity compared with the
economies of East Asia. It also has weak institutions, including poor
regulatory effectiveness and high levels of corruption.
Second, one of the reasons that Mexico did poorly in competition
with China was that China was investing heavily in infrastructure
and education. Mexico’s limited tax revenues, exacerbated by the
loss of tariff revenue, was one reason why it did not make the neces-
sary investments.
Third, NAFTA was not really a free trade agreement. America

retained its agricultural subsidies. NAFTA pitted the heavily sub-
sidized US agribusiness sector against peasant producers and family
farms in Mexico. US farmers export many of their products into
Mexico at costs far below those of the local market, driving down
prices for local farmers. Moreover, America continued to use what
were effectively non-tariff barriers to keep out some of Mexico’s prod-
ucts. These policies had deleterious effects on rural livelihoods. One-
fifth of Mexico’s workers are employed in the agricultural sector and
75 per cent of Mexico’s poverty is found in rural areas. While some
large Mexican agribusiness sectors have expanded their exports, much
of Mexico’s rural sector is in crisis. Local farms are threatened by
cheap imports from the US, falling commodity prices, and reduced
government support. Four-fifths of the population of rural Mexico
lives in poverty, and more than half are in extreme poverty.
Mexico’s experience with NAFTA provides a cautionary tale. The
goal of economic integration should be to raise living standards, but
it is clear that trade liberalization by itself is not sufficient to achieve
this. There is no doubt that trade and investment are vitally import-
ant for economic growth but the real challenge is to pursue liberal-
ization in a manner which promotes sustainable development, in
which those at the bottom and middle see incomes rising.
Trade liberalization, welfare, and growth
The intuition behind the notion that trade is welfare-enhancing is

simple. Imagine two people exchanging goods with each other. They
would voluntarily trade their goods if and only if they both benefit
from doing so. Thus government intervention to prohibit, restrict, or
tax their trade restricts their ability to realize welfare gains from
such mutually beneficial exchange.
However, trade among countries is somewhat more complex.
In the basic economic model, trade is beneficial because it allows
each country to specialize in the goods that they produce relatively
efficiently. This principle of ‘comparative advantage’, established by

the nineteenth-century economist David Ricardo, is the core of trade
theory and is the foundation of its normative implication in favour
of free trade.
In addition to the gains from specialization according to comparat-
ive advantage, trade may deliver benefits and costs through four
other channels. Trade liberalization opens foreign markets, expand-
ing the demand for domestic firms’ goods and enabling them to serve
a larger market and realize gains from economies of scale. Trade lib-
eralization may make available a range of inputs at lower prices, low-
ering costs of production. Liberalization may also introduce more
competition from foreign firms to the domestic economy, which
may result in improvements to the efficiency of local production.
Finally, trade liberalization may, through various channels, affect
the rate of economic growth.
Most of the traditional arguments for free trade are, however, based
not on growth but on efficiency, i.e. liberalization leads to a change in
the level of welfare rather than any change in the long-run rate of
growth. The basic results were formalized in modern economics by
Paul Samuelson (1938), who showed that free trade is superior to
autarky, and later (1962) that it is also superior to any intermediate
regime of trade restrictions.
However, the underlying assumptions which yield that conclu-
sion are highly restrictive, and often fail to capture relevant features
of developing countries’ economies. The standard argument in favor
of trade liberalization is that it improves the average efficiency in a
country. Imports from foreign producers may destroy some ineffi-
cient local industries, but competitive local industries are supposed
to be able to absorb the slack as they expand their exports to foreign
markets. In this way, trade liberalization is supposed to allow
resources to be redeployed from low-productivity protected sectors
into high-productivity export sectors. But that argument assumes
that resources will be fully employed in the first place, whereas in
most developing countries unemployment is persistently high.
One does not need to redeploy resources to put more resources into
the export sector; one simply needs to employ hitherto unused
resources. In practice trade liberalization often harms competing
local import industries, while local exporters may not automatically

have the necessary supply capacity to expand. So liberalization often
seems to result in labour temporarily going from low-productivity
protected sectors to zero-productivity unemployment. Unfortunately,
most of the models which attempt to address questions of welfare
gains from trade liberalization assume full employment, and there-
fore provide no answers to this key question: the impact of liberal-
ization in economies with underemployed resources. But the issue
of unemployment is not just a theoretical problem. Concern that
trade liberalization will lead to increased unemployment is perhaps
the most important source of opposition to liberalization. And the
concerns are particularly relevant in countries where there is no
unemployment insurance and weak social safety nets.
A second assumption of the model underlying the conclusion that
trade liberalization is welfare-enhancing is the existence of perfect
risk markets. But there is high volatility in international markets, risk
markets are highly imperfect, and trade policy can reduce exposure to
risk (Dasgupta and Stiglitz 1977). Without trade, producers are insu-
lated from the full force of output fluctuations by built-in insurance: if
there is a reduction in the quantity of output firms can produce, then
they can charge a higher price for it. Thus their incomes vary less than
output. Trade may weaken this automatic insurance, since in small
economies prices will be determined on the world market and will be
unrelated to domestic output. Because their incomes will be more
variable, risk-averse firms will invest less in some sectors with high
returns but high variability; and as the economy moves into lower-
return, less variable activities, total output will decline. Under quite
plausible conditions, one can show that free trade is Pareto-inferior to
autarky—everyone is worse off—which is just the opposite result to
that of the conventional wisdom (Newbery and Stiglitz 1984). These
market failures underline the importance of sequencing. It makes a
great deal of difference, for instance, whether trade liberalization
occurs before or after risk markets or social insurance programs have
been created. But imperfections of risk markets are inherent, a conse-
quence of imperfect and costly information.
One of the strengths of the market economy is that prices provide
all the coordination that is required, i.e. there is no need for a central
planner. But in developing countries, markets are often absent or,

even when present, often do not work well, and prices accordingly
are not able to perform this critical function. The problem, in some
sense, is intrinsic. The process of economic development entails the
creation of whole new industries. But some industries are dependent
on inputs from other, ‘intermediate’ industries, for example, the car
industry is dependent on the steel industry for fabricated metals.
Intermediate goods industries will not be created until the final
goods industries that use those intermediate goods are created; but
the final goods industries cannot be created until the intermediate
goods that they require are available. It is too much to expect mar-
kets for goods not yet produced to function well! Curiously, these
arguments have been used both to support and to criticize trade
interventions. In one sense, trade helps countries get around the
need for planning. A country does not need to develop its own inter-
mediate goods industries if it can import intermediate goods. In
addition an open developing country does not need to rely on its own
local demand. It can take advantage of the global market to attain the
requisite economies of scale in tradeable goods. However, many of
the key intermediate inputs are non-tradeables, so there is still a need
for coordination, especially if there are significant scale economies
in the non-tradeables. This, in turn, provides one of the critical argu-
ments for trade restrictions: to get the necessary scale, one may have
to restrict competition from foreign producers.1 The existence of
these market failures suggests a need for government intervention.
The appropriate form for the intervention needs to take into account
limitations on the information available to government and the
nature of the (irremediable) market failures.
The need for government revenue can also, in some circumstances,
provide a rationale for trade taxes. One corollary of the classic
Diamond–Mirrlees production efficiency theorem (1971) is that it is
optimal in a small open economy for the government to raise revenue
from tax on the net demand of households rather than from border
taxes (see Dixit and Norman 1980). But the conditions under which
this is true are very restrictive, and especially do not hold in most
1
Especially, as suggested above, in the presence of capital market imperfections.
developing countries (Dasgupta and Stiglitz 1972). In many develop-

ing countries, moreover, tariffs are the main source of government
revenue, and where this is the case, the optimal tariff structure may
not be uniform (Dasgupta and Stiglitz 1974). Max Corden (1974)
argues that, particularly in developing countries, the collection costs
associated with trade taxes are likely to be much smaller than those of
income and commodity taxes. Where this is true, trade taxes might be
the best revenue-raising device. Recently international institutions
have been encouraging developing countries to reduce their trade
taxes in favour of indirect commodity taxation such as a value-added
tax (VAT). However, many developing countries have large informal
sectors which are beyond the reach of indirect taxation. In these cir-
cumstances, M. Shahe Emran and Joseph Stiglitz (2004) have shown
that a switch from trade to indirect taxes may be welfare-reducing.
Trade liberalization will also affect inequality. Opening up to trade
does not make everyone in a country better off. Instead it changes the
distribution of income and creates winners and losers. The standard
economic argument is that the net gains from trade liberalization are
positive so the gainers can compensate the losers and leave the coun-
try better off overall. Unfortunately, such compensation seldom
occurs. These distributional consequences are an important pract-
ical consideration. They provide much of the political opposition to
trade liberalization. And they become more salient in global inter-
national trade regimes which are viewed to be ‘unfair’.
Moreover the standard theory assumes that taxes and subsidies to
compensate the ‘losers’ from trade liberalization are costless. But there
may be large inefficiencies associated with redistributive schemes.
Once the distributional consequences are taken into account, trade lib-
eralization may not be Pareto-superior; it may not be possible to make
everyone better off. Concerns about inequality may place a limit on the
desirability of liberalization in situations, especially where a market
distortion creates a link between the distribution and allocation of
resources. Sudhir Anand and Vijay Joshi (1979) describe the situation
(fairly common in developing countries) where workers in the urban
sector receive a higher wage than those in the rural sector,2 giving the
2
While in their model the wage differential is simply given exogenously, efficiency wage theory (e.g.
Shapiro and Stiglitz 1984) provides an explanation for such differentials.
government an incentive to intervene to increase employment in

the urban industries. Anand and Joshi show that, under fairly general
conditions, a Pareto-efficient outcome cannot be achieved using
wage subsidies alone. In their model the rigid factor price creates an
unavoidable trade-off between efficiency and equality even when
the standard best-policy instrument is available and can be financed
by lump-sum taxation. They conclude that ‘[d]epartures from tech-
nical efficiency may be called for as part of the rational response by
governments to the limitations they face in carrying out desirable
income distribution policies’.
Ricardo Hausmann and Dani Rodrik (2004) demonstrate another
type of market failure based on information externalities. They
emphasize the importance of entrepreneurship in developing coun-
tries. These entrepreneurs are engaged in a process of ‘discovering’
which economic activities will be successful in their country.
Hausmann and Rodrik point out that successful (and failed) entre-
preneurs provide information to the market which is of great social
value yet is poorly remunerated. If the entrepreneur fails in his ven-
ture, he bears the full cost of the failure: if he is successful he shares
the discovery with others who enter the new industry. ‘The entre-
preneurs who figured out that Colombia was good terrain for cut
flowers, Bangladesh for t-shirts, Pakistan for soccer balls, and India
for software generated large social gains for their economies’, but
could keep very few of these gains to themselves. Thus ‘it is no great
surprise that low-income countries are not teeming with entrepre-
neurs engaged in self-discovery’ (Rodrik 2004). Coordination failures
may occur when the profitability of new industries depends on the
simultaneous development of upstream and downstream industries,
or when new industries are characterized by scale economies and
have non-tradeable inputs (Rodrik 1996).
While most of the economic theory of trade liberalization has
focused on static welfare gains, the long-term effects of trade liberal-
ization are determined by its effect on the economy’s rate of growth.
Recent models of endogenous growth have important implications
for the theoretical relationship between free trade and economic
growth, although their results are not fully understood, and their
policy consequences remain to be thoroughly established.
There are several possible arguments for why increased trade may
lead to faster sustained growth rates. One argument is that larger
markets will lead to larger returns to investments in R & D. This
would suggest that a global regime with free trade will be associated
with higher overall growth rates. A similar argument is that, with
learning by doing, there are strong spillovers within a country (i.e. the
learning in one firm spills over to other firms) but not across
countries. Here, accordingly, the pace of innovation is related to the
scale of production within the country: if there is more specialization,
there will be more learning. A third argument focuses on the fact,
again noted earlier, that with a larger market, there can be a larger
variety of inputs, which can sustain not only more efficient produc-
tion but a faster pace of innovation.
These theories suggest the importance of knowledge, learning,
and human capital. One possible implication for policy (though not
necessarily trade policy) is that countries may benefit by promoting
more technologically dynamic sectors. Research by Dani Rodrik and
others suggests that developing countries may need to ‘embed pri-
vate initiative in a framework of public action that encourages
restructuring, diversification, and technological dynamism beyond
what market forces on their own would generate’ (Rodrik 2004). For
example, the theory of comparative advantage told South Korea, as it
emerged from the Korean War, that it should specialize in rice. But
Korea believed that even if it were successful in increasing the pro-
ductivity of its rice farmers, it would never become a middle- or
high-income country if it followed that course. It had to change its
comparative advantage, by acquiring technology and skills. It had to
focus not on its comparative advantage today, but on its long run, its
dynamic comparative advantage. And government intervention was
required if it was to change its comparative advantage.
The existence of these market failures or learning externalities
may suggest a need for government intervention, but they do not, by
themselves, justify trade policy as the best instrument. Research
into policy choices has been developed into a general theory of dis-
tortions (see Corden 1957) which generally show that tariffs and
other introduced trade distortions are usually the nth-best policy
instrument, after various types of subsidies including training,
employment, output, and knowledge diffusion subsidies (if those are

available). But often, in poor and backward developing countries,
these instruments may not be immediately practical or affordable.
This underlines the importance of sequencing. Even if trade policy is
the nth best instrument, trade liberalization should not occur until
one of the n ⫺ 1 preferred alternatives is feasible and successfully
implemented.
However, in many cases the market failure might be endemic or
not readily able to be countered by government action. Much of the
earlier literature on correcting market distortions made the critical
error of treating the distortion almost as if it was an accidental
mistake, which government could easily correct. For example, if
markets were imperfect because wages were rigid, the solution was
to make wages more flexible, i.e. if the reason for the rigidity was
government minimum wages, there was an obvious answer—
eliminate the minimum wage. To take another example, if there
were capital market failures, the solution was to create perfect capi-
tal markets. These direct interventions to correct market failures
were preferred to indirect government action, including trade and
industrial policies.
But both wage rigidities and capital market imperfections can
arise from information imperfections (Stiglitz 2002) and these might
be difficult to correct with direct action. So long as it is costly to
acquire information, there is no easy way of eliminating these mar-
ket imperfections. To be sure, government too faces information
(and other) constraints; but the nature of those constraints, as well as
the objectives, differs between the government and private firms.
Optimal interventions may thus involve trade and industrial policies,
as the discussion below makes clear.
Consider, for instance, the infant industry argument for protec-
tion presented earlier. Critics of the infant industry argument con-
tend that if a firm will eventually become profitable, then it should
be able to finance its learning phase through private capital markets,
and, accordingly, if the benefits of this learning stay wholly within
the firm then there is no case for government intervention. Only
some imperfection in the capital market justifies government
action, and even then, the best policy, if available to developing
countries, would be to improve the capital market rather than to

impose trade distortions.
But the modern theory of asymmetric information explains why
capital market imperfections are inherent, and not just a happen-
stance, and why governments cannot simply remedy these capital
market imperfections (Stiglitz and Weiss 1981). These capital mar-
ket imperfections are particularly relevant here, for banks would
have to be willing to lend to enable firms to sell below cost, in the
hope that by doing so their productivity will increase so much that
they will become a viable competitor. It should be obvious that such
loans would be viewed, at best, as highly risky. (And banks would
likely view themselves at a marked informational disadvantage in
judging whether the firm’s claims about its future prospects—in
spite of large current losses—have credibility.)
Once capital market imperfections are taken into account, then
protection may be optimal, as Dasgupta and Stiglitz (1980) show.
They argue, in particular, that protection may have advantages over
other instruments, e.g. subsidies, when government has limited
sources of income and limited ability to target. Interestingly, these
are among the reasons that some argue for private production of
knowledge, through intellectual property rights, as opposed to pub-
lic production, through direct government subsidies. The govern-
ment cannot identify who will be good producers of knowledge, and
therefore cannot target the subsidies; the market is a self-selection
mechanism. Intellectual property rights increase the returns to pri-
vate production of knowledge, at a cost—temporary monopoly rights.
Protectionism limits competition from abroad, but allows competi-
tion from within. By increasing the private returns, it helps to better
align private and social returns to innovation. This is even more rel-
evant in the infant economy argument for protection (where learn-
ing spillovers extend beyond the industry to the entire economy.)
Theoretical arguments which caution against a full embrace of
free trade abound. But the question for policy-makers is not whether
these arguments exist, but whether they carry enough weight to be
acted upon. In a thoughtful article, two noted proponents of free trade
for developing countries, T. N. Srinivasan and Jagdish Bhagwati
(1999), review some of the above theory and many other theoretical
arguments which caution against the universality of the benefits of

free trade. They concede that it is possible to build many theoretical
models in which free trade will ‘reduce current income and even
growth . . . if market failures are present’. But they challenge inter-
preters of this evidence to ask themselves the following question: in
formulating policy, ‘do we view these models as representing a “cen-
tral tendency” in the real world or merely “pathologies”?’, and they
caution policy-makers not to become prisoners of the nihilistic view
that ‘because anything can be logically shown, nothing can be empir-
ically believed and acted upon’. In order to determine whether the
case against unconditional trade liberalization for developing coun-
tries is important or merely a series of inconsequential theoretical
possibilities, it is necessary to turn to the empirical evidence.
Empirical evidence
A number of studies have attempted to show that there is a system-

atic relationship between growth and trade and/or trade liberaliza-
tion, using cross-country studies. The hypothesis has been that,
holding other things constant, countries that have liberalized more
or which trade more have grown more.
Our previous discussion suggests that only under certain circum-
stances will that be the case. Given the complex and contingent rela-
tionship between trade liberalization and economic growth, and the
manifold difficulties associated with empirically testing this rela-
tionship (discussed briefly below), it comes as no surprise that most
economists, even most of those that have no reservations about lib-
eralization, accept that the empirical literature has been inconclus-
ive (Winters 2003). The economic growth literature has been
successful in demonstrating the importance of some variables for
economic development, including education, institutions, health,
and geography. However, the relationship between trade liberaliza-
tion and growth is much more controversial.
The weakness of the evidence in favour of a direct relationship
between trade liberalization and economic growth has not prevented
some economists from pursuing free trade at full throttle. The IMF’s
former First Deputy Managing Director, Stan Fischer (2000), boasted

that the ‘Fund is a powerful voice and actor for free trade’ and sug-
gested that this is because ‘integration into the world economy is the
best way for countries to grow’. The IMF may be right to promote lib-
eralization (by developed and developing countries alike), though
that is hardly its mandate, and it would do better focusing on trying
to enhance global financial stability. But it should be pointed out
that the empirical evidence that it even has a positive effect on
growth is mixed, that it almost certainly is not the most important
factor in growth, that the theory suggests important caveats, and
that the experience of successful countries indicates that the reform
process should be managed gradually and carefully. Integration
through exports, as in East Asia, has a far more convincing record
than integration through rapid liberalization. In short, trade liberal-
ization should be a tailored policy, not a one-size-fits-all.
It is difficult to identify the evidentiary source of the bullishness
for unqualified trade liberalization. Certainly there were several
studies in the early 1990s which purported to show a positive rela-
tionship between trade openness and economic growth (see Dollar
1992; Ben-David 1993; Sachs and Warner 1995), but even these were
careful to qualify their results. In the conclusion to their paper,
Jeffrey Sachs and Andrew Warner point out several of the important
caveats to their study. Their studies focus on trade, not on trade lib-
eralization. Francisco Rodriguez and Dani Rodrik (1999) have per-
suasively shown that the conclusions of these studies should be
interpreted with extreme caution. They found that the indices of
openness used in these studies conflated the effects of trade policies
with other phenomena. In particular the studies were identifying the
negative effects of macroeconomic imbalances, instability, and geo-
graphic location, and misattributing them to trade restrictions.
Rodriguez and Rodrik pointed out that because of these method-
ological weaknesses, the policy conclusions drawn from these stud-
ies are not strongly supported by the data they present.
To recognize the weakness of the empirical evidence in this field is
not to argue that trade protection is good for growth. Rodriguez and
Rodrik themselves point out that there is no credible evidence that
trade restrictions are systematically associated with higher growth
rates in the post-war period. But it does suggest that the relationship
between trade liberalization and growth is not simple. Preliminary
research being conducted at Columbia University, for instance, sug-
gests that trade liberalization may have positive effects on countries
with low unemployment rates, but negative effects on countries with
high unemployment rates. More generally, the empirical evidence
supports the view that the benefits of trade liberalization depend on a
range of other factors which are difficult to observe separately because
of measurement problems3 and other econometric difficulties.4
Policy implications
Theory and empirical evidence indicate that trade liberalization can be

a positive force for development in poor countries, but that these bene-
fits depend on other, concomitant factors. Given this, we would hope
the focus of current economic research to be on how differences across
countries affect their experience of liberalization, and we would hope
the focus of policy research to be on how trade policies can best be
tailored to the particular circumstances of different countries.
But we would hope for a consensus of answers among policy-
makers, but we would hope for a consensus of approach. However,
international trade negotiations exhibit no such consensus—the
acrimonious breakdown of talks at Cancún and Seattle and the ongo-
ing polarization among academics, NGOs, and international insti-
tutions are testament to that. There are still those on the right
who would press developing countries to move immediately and
uncompromisingly towards free trade. And there are still those on
the left who believe that the way to help developing countries is to
shield them vigorously from the forces of reform and liberalization.
3
For instance, many of the empirical studies treat trade regimes as a binary variable (i.e. countries are
either ‘open’ or ‘closed’), which ignores the subtlety and dynamics of different trade regimes. Studies which
analyse trade-weighted average tariff rates will underweight the importance of high tariffs because the
quantity of imports in those tariff lines is likely to be low. Many studies ignore non-tariff barriers, and those
that include them have a hard time distinguishing which are important and which are not.
4
For instance, many of the indices used to analyse openness may be endogenously related to other pol-
icy or institutional variables which have an independent negative effect on economic growth, making these
indices inappropriate variables with which to analyse the direct effect of trade liberalization on growth.
The theoretical and empirical evidence may not speak clearly on all
issues, but it certainly rules out the extreme positions on both sides. It
is therefore worth asking why these extreme positions have proved so
enduring. On the left, the fault probably lies with overzealous altru-
ism. The anti-globalization movement has done much to raise aware-
ness about important international issues, but as is often the case with
unstructured social movements, their public message has sometimes
become distorted. Unfortunately those within the movement who
can attract the most publicity are not always those with the strongest
analysis. The unfortunate message from the hard-line activists is that
a good round of trade negotiations is one that requires the developing
countries to do nothing in the way of reform. They encourage devel-
oping countries to look outward to the developed countries as the
primary source of and solution to their problems. Unfortunately,
unreasonable demands for “reforms” from the developing countries,
unsupported by or in some cases counter to historical experiences,
strong empirical evidence and theory—reforms which might in fact
set development programs back—enhance support for these posi-
tions, especially reforms demanded by trade negotiators who other-
wise have evidenced little real concern for the developing countries’
welfare; this is true even where it is recognized that liberalization of
developed countries’ trade policies is not a substitute for, but rather
than a complement to, developing countries’ own internal reforms.
There is a tendency among some in the anti-globalization move-
ment to view the extreme position taken on trade liberalization by
many on the right as a revealing testament of their malevolence or
wilful disregard for the problems of the world’s poorest countries. Of
course, at least on the part of the academics who make serious con-
tributions to the debate, there is no such malevolence or disregard.
How then can they continue to insist, in the face of the theoretical
and empirical evidence, that developing countries pursue rapid and
unfettered trade liberalization? One answer is that they are more
concerned about government failure than market failure.
Many economists have serious reservations about the ability of
officials from developing countries to manage anything but the sim-
plest, most liberal trade policy regime. Alan Winters (2003), Director
of the Development Research Group at the World Bank, notes that
‘[t]here are undoubtedly hundreds of individual cases where a one-off

policy intervention would be beneficial, for example where protec-
tion would allow learning or training, or generate a terms-of-trade
gain, or support a poor family while it learned new skills.’ But he
rejects the use of such second-best interventions to overcome mar-
ket failure because, among other things, he believes that developing
country officials are not skilled enough to identify and implement
them effectively. He says, ‘the application of second-best economics
needs first-best economists, not its usual complement of third- and
fourth-raters’. The same argument was espoused by Anne Krueger,
the First Deputy Managing Director of the IMF, in her presidential
address to the American Economic Association in 1997. She
claimed, ‘most [trade] policy implementation is carried out by
government officials . . . [who lack] . . . the degree of sophistication
needed to interpret research results’ (Krueger 1997).
This reason for adopting and espousing the simple orthodoxy of
free trade is not that such policies are optimal in a standard economic
sense, but that any more complicated development strategy would be
beyond the capability of the officials of developing countries.
There are many reasons why government interference in the
economy may fail. One is that governments are not possessed of
full information about market failures; indeed they will often have
less information than private agents about particular sectors. This
impedes the identification of both the market failure and the (poten-
tially unintended) side effects of any proposed solution. It may be
possible to overcome these information problems, but this may be
costly, and the government needs to insure that the net gain is larger
than for alternative policies. There are many examples of failed trade
and industry policy experiments. But, as Dani Rodrik points out, some
failures are to be expected. All that needs to be true is that the govern-
ment intervention delivers social returns on average across a range of
projects within each policy. If we observed no failures, the policy
would arguably be being carried out in too timid a fashion.
Another important concern linked to government failure is that
simple trade regimes are more transparent and less prone to corrup-
tion and rent-seeking activities. Less distorting policies usually offer
fewer opportunities for corruption.
In our view, the fear of government failure is real, but it is not over-
riding. Putting aside the condescending attitude towards developing
countries evidenced in assertions about ‘third- and fourth-raters’,
the fact is that some developing countries—most notably in East
Asia—have arguably been extremely successful in managing trade
and industrial policies to promote growth. Concerns about govern-
ment failures may provide a reason that such policies may be
eschewed in the long run, but they do not trump all development
objectives, market failures, and adjustment costs in the short run.
Policies should be designed to minimize the risk of government fail-
ure, for example through carefully designed institutions and policies
(see Rodrik 2004) and international technical assistance. The
remaining risk should be appropriately weighted within the policy-
makers’ decision-making process.
None of today’s rich countries developed by simply opening them-
selves to foreign trade. As Ha-Joon Chang (2001) has documented, all
the developed countries used a wide range of trade policy instru-
ments which should make their WTO ambassadors blush when they
sit down to negotiate with today’s developing countries. Chang’s
evidence does not prove that interventionist trade policies were, or
are now, the best policies for development, but it does show, at the
very minimum, that the risk of government failure can be managed
in countries as they develop.
China and India provide more proximate evidence of this fact.
Both have successfully integrated into the world trading system, and
both have benefited greatly from international trade, yet neither fol-
lowed orthodox trade and industrial policies. China has been partic-
ularly careful to ensure that its economic development strategy is
gradually implemented and carefully sequenced. Certainly China
has become more open in recent years, and has benefited from doing
so, but trade liberalization certainly did not cause China’s growth.
China began to grow rapidly in the late 1970s, but trade liberaliza-
tion did not start until the late 1980s, and only took off in the 1990s
after economic growth had increased markedly. The Indian story is
similar: growth increased in the early 1980s while tariffs were actu-
ally going up in some areas and did not begin to come down signifi-
cantly until the major reforms of 1991–3.
The way forward

Theory teaches us that when markets are perfect, trade-distorting
policies will be welfare-reducing, and even when markets suffer from
distortions, trade policies may not be the best instruments to over-
come them. But, as we have seen, both theory and evidence provide
less guidance about policies in a real world marked by market imper-
fections, many of which are intrinsic, arising whenever information
is incomplete and costly and there are costs associated with creating
markets.
These concerns are of particular relevance to developing countries.
Developing countries certainly do not have perfect markets. Many of
their markets are missing or incomplete, particularly markets for
insurance and credit. Public goods are undersupplied, coordination
failures are rife, and the social benefits of entrepreneurship are larger
than the expected private returns. The adjustment costs associated
with liberalization would be large and exacerbated by high unem-
ployment and weak social safety nets.
Often there will be better instruments than trade policy with
which to overcome these market failures and soften the adjustment
costs associated with reform. But poor governments with small pub-
lic resources have a limited number of instruments at their disposal.
Often it would be a mistake for these governments to liberalize their
trade regime before they have put compensating policies in place.
Thus, for developing countries, we can put aside the debate about
whether the world should move towards free trade. For many of
them, the issue is largely about which steps towards liberalization
make sense, and pacing and sequencing: what should be done before
they liberalize, and how fast the liberalization agenda should be
pushed.
There is a middle ground between the extreme positions of the free-
traders and the anti-globalizers. This middle ground recognizes that
even if one accepts the ultimate desirability of free trade, rushed
liberalization may be harmful. Policies in the middle ground need to
be found by investigating the effects of market failures on the experi-
ence of liberalization in different countries. Developing countries
should attempt to promote development by correcting these market
failures through policy interventions, including trade policies, if, and

only if, they are the best available instruments. Policy-makers should
recognize the potential for government failure arising from their
interventions. They should neither ignore this risk nor fear it. Instead
they should look for ways to overcome it and, where those are not
apparent, appropriately weight the risk in their policy-selection
process.
Developed countries must do their part. They can help to integrate
developing countries into the world trading system and ensure that
they benefit from it. As we explain later in the book, developed coun-
tries need to reform their own trade policies in ways that open trad-
ing opportunities for the developing countries.
The developed countries play the central role in the politics of
global trade negotiations and are responsible for much of what hap-
pens in the WTO. The developed countries have a responsibility to
build the global trade architecture in ways that enhance the participa-
tion of the developing countries.
In this book we suggest a policy program within the WTO which
would benefit the developing countries. Developed countries are a
crucial part of this program because so much of world trade is
affected by their policies and because they are the most powerful
actors within the WTO. But developed countries are neither the
whole problem nor the whole solution. Their trade policies are
important, but their reform is a complement to, not a substitute for,
reform within developing countries.
3
The Need for a
Development Round
3
Developing countries and world trade
The current multilateral trade liberalization round, the Doha

Round, is the ninth in a series of such negotiations which began in
Geneva in 1947. The first eight of these were conducted under the
auspices of the WTO’s predecessor, the GATT (General Agreement
on Tariffs and Trade), which was established on a provisional basis
after the Second World War as a draft charter for the proposed
International Trade Organization (ITO). The ITO was stillborn—
it was never ratified by the US Congress and other national
legislatures—but the GATT continued to govern international trade
in the form of a multilateral treaty from 1948 to the creation of the
WTO in 1995.
The GATT set forth the principles under which the signatories, on
a basis of reciprocity and mutual advantage, would negotiate a sub-
stantial reduction in customs tariffs and other impediments to trade,
and the elimination of discriminatory practices in international
trade. As more countries joined, the GATT became a charter govern-
ing almost all world trade except for that of the communist coun-
tries. Despite its inauspicious beginnings and its provisional nature,
the GATT was successful in promoting the liberalization of a con-
siderable proportion of trade and contributed to the substantial
Table 3.1. The nine trade negotiation rounds under the GATT and WTO
Year Place/Name Subjects covered Number of countries
1947–8 Geneva Tariffs 23

1949 Annecy Tariffs 13
1950–1 Torquay Tariffs 38
1956 Geneva Tariffs 26
1960–2 Dillon Round Tariffs 26
1963–7 Kennedy Round Tariffs and anti-dumping 62
measures
1973–9 Tokyo Round Tariffs, non-tariff barriers 102
(NTBs), ‘framework’
agreements
1986–3 Uruguay Round Tariffs, NTBs, rules, services, 123
intellectual property, dispute
settlement, textiles,
agriculture, creation of WTO
2001– Doha Round (Under negotiation) 142⫹
growth of world trade throughout the post-war era.1 Through its

principles of transparency, non-discrimination, and reciprocity, the
GATT gave predictability to world trade and provided a reasonably
open forum for the mutual exchange of liberalization concessions
and the settlement of disputes for almost half a century (Table 3.1).2
Developing countries played only a small role in the seven rounds
of trade negotiations in the first 20 years of the GATT.3 To the extent
that they participated at all, they campaigned for special treatment.
This took the form of preferential access to the rich countries’ mar-
kets at tariff rates below those applied to other countries (eventually
enshrined in the Generalized System of Preferences) and exemptions
from GATT rules. Article XVIII in the GATT rules provided devel-
oping countries with differential treatment. Among other excep-
tions, it allowed economies ‘which can only support low standards
of living and are in the early stages of development’ to ‘implement
1
Trade growth was consistently larger than production growth throughout the GATT era.
2
For a history of the political economy of the GATT see Hoekman and Kostecki (1997).
3
Developing countries hardly participated at all in the substantive agreements of the Kennedy Round.
More than 70 developing countries participated in the Tokyo Round, where the enabling clause was
adopted which introduced the concept of special and differential treatment (S&D), but their participation in
broader negotiations was limited.
THE NEED FOR A DEVELOPMENT ROUND 43
programmes and policies of economic development designed to raise

the general standard of living of their people, to take protective or
other measures affecting imports’. This recognized the right of
developing countries to impose quantitative and other restrictions
to protect their infant industries. Because of Article XVIII, develop-
ing countries could simultaneously be members of the GATT and
evade the obligations imposed on developed countries.
This special status gave developing countries more freedom in
determining their own development policies, but it simultaneously
allowed them to be marginalized within the substantive negotia-
tions. Even in the absence of their special status, tariff negotiations
do not offer many incentives for smaller economies to participate
actively, because they have little to offer large markets in reciprocal
negotiations. But when exemptions freed developing countries from
the requirement of reciprocity, their ‘commitments’ became almost
meaningless because they were at liberty under the special arrange-
ments to raise tariffs or introduce other protection at their own
discretion. They therefore had relatively little bargaining power
within the GATT rounds because their concessions were of doubtful
value. They had no ‘skin in the game’.
A passive strategy within the GATT negotiations actually
suited some of the largest developing countries (especially China,
India, and Brazil) which, until the 1980s, did not see international
trade as their primary engine of growth. Instead they pursued pol-
icies of ‘import substitution’ which isolated their economies from
the rest of the world. These countries sought economic develop-
ment from internal sources and they had only a small incentive to
pursue progress on multilateral liberalization within the GATT
regime.
The developed countries were far more interested in each others’
markets than in those of developing countries. Bergsten (1998)
describes the history of the post-war multilateral trading system
in terms of the power relationships between the rich countries,
particularly the EU and US. He argues that the initial creation of the
European Common Market in the late 1950s was one of the key
motivations for the American initiative to launch the Kennedy
Round in the early 1960s. The US wanted to reduce the newly
created discrimination against American exports, and also to build

the ‘new Atlantic partnership’ enunciated by President Kennedy.4
Similarly, the expansion of the European Community to include the
United Kingdom and others, with the extension of its discrimination
to important new markets, was an important factor in the American
decision to insist on the Tokyo Round in the 1970s.5 Thus the prim-
ary concern of developed countries was to gain access to other devel-
oped markets and the reform agenda focused on efforts to liberalize
those goods traded intensively between developed countries.
Together these factors explain the peripheral role of the develop-
ing countries in most of the GATT negotiations. As a consequence,
over several rounds of agreements, the world trading system was
tailored to the interests of the developed countries, who had, particip-
ated intensively in the negotiations. Protection was progressively
reduced on the goods of export interest to developed countries, but
remained on goods exported intensively by developing countries. It
was no surprise that the GATT came to cover trade in all goods
except agriculture and textiles—both of interest to developing
countries—which were the subject of separate agreements whose
provisions were far less liberal than those of the GATT. Textiles were
covered by the Multifiber Arrangement (MFA), through which
developing countries bargained bilaterally to establish quotas on the
quantities of exports that they could export to developed countries.
Only developing countries were discriminated against: developed
countries did not impose any restrictions on textile imports from
other developed countries. Similarly agricultural trade was excluded
from the GATT and developed countries still continue to pursue pro-
tectionist agricultural policies which distort trade and harm pro-
ducers in developing countries. As early as the mid-1960s, the world
trade system was recognized as being unfair: ‘In an important sense
the trade policies of the developed countries may be said to discrim-
inate against the less developed countries. While they generally do
4
The Kennedy Round also originated partially in a monetary crisis (see Bergsten 1998). President
Kennedy reportedly worried about the balance-of-payments problem and saw access to foreign markets for
American exports as a potential solution.
5
Similarly, the EU decision to launch the ‘single market’ in 1985, with the implied broadening of dis-
crimination to many new types of economic activity, also added to the US desire to begin the Uruguay
Round.
not discriminate against those countries in the form proscribed by

the most-favored-nation principle, . . . their policies are in effect dis-
criminatory in that the most serious barriers are erected in goods
which the less developed countries typically have a comparative
advantage in producing—agricultural commodities in raw or
processed form, and labor-intensive, technologically unsophistic-
ated consumer goods’ (Johnson 1967: 79).
In the 1980s developing countries came to play a larger role in
international trade policy. Each of the obstacles to their participa-
tion had been reduced. First, the increased importance of developing
countries’ trade—especially the newly industrialized countries in
East Asia—increased the developed countries’ interest in their mar-
kets and led to calls for them to stop ‘free riding’ and enter the sys-
tem. At the same time developing countries placed more stock on
the development potential of trade as many of them turned away
from import substitution policies. As they did so, policy-makers in
newly opened developing countries recognized the importance of
participating in trade negotiations. As a result, in contrast to the
Kennedy and Tokyo Rounds, developing countries were actively
involved in the discussions that led to the Uruguay Round: the large
number of accessions or requests for accession to the GATT from
such countries at that time (including Mexico and China) indicated
that the issues on the table were being taken seriously by a growing
number of developing countries.
The Uruguay Round marked a shift in the history of trade negotia-
tions. The largest industrial countries, led by the US, attempted to
extend the GATT system to cover services as well as new areas of
domestic policy which were deemed to be ‘trade-related’. Despite ini-
tial resistance from some developing countries, an agenda was agreed
at the Uruguayan resort city of Punta del Este in September 1986. The
agreement was that developing countries would negotiate on the new
issues of services, Trade-Related Aspects of Intellectual Property
Rights (TRIPS), and Trade-Related Investment Measures (TRIMS). In
return, they would get better market access for their exports of goods.
The Uruguay Round, for all its faults and imbalances, was the most
ambitious set of trade negotiations ever. It covered reductions in tar-
iff and non-tariff barriers in industrial and agricultural goods, as well
as textiles and clothing, and it attempted to extend multilateral rules

to new areas, notably services and intellectual property.
Redressing past imbalances
In 1993 the Uruguay Round was finally being brought to a close. Part
of the impetus for members to conclude the round was the promise
of large welfare gains that had been projected by many researchers
(See Table 3.2). In 1992–3, the World Bank, the US Organization
for Economic Cooperation and Development, and various other
institutions made projections of welfare gains on the order of
US$ 200 billion a year.6 A large share of the gains was predicted to
accrue to developing countries.7
In hindsight these estimates—particularly in relation to develop-
ing countries—were somewhat over-optimistic. It has since been
estimated that the vast majority of the gains from the Uruguay Round
would accrue to developed countries, with most of the rest going to
a relatively few large export-oriented developing countries. Indeed
Table 3.2. Early estimates of income gains from the Uruguay Round Agreement ($bn)
Study World Developing countries
Harrison, Rutherford, and Tarr (1995) 52.5–188.1 4.8–61.7

Francois, McDonald & Nordstrom (1995) 51.4–251.1 9.0–91.9
GATT Secretariat (1993) 230 65
World Bank (1993) 213 78
Nguyen, Perroni, and Wigle (1993) 212.1 36
OECD (1993) 274.1 89.1
Deardorff (1994) 140–260 —
Source: Epstein (1995).
6
Indeed after Marrakech, the GATT Secretariat put forward a larger estimate of the minimum gain—
US$500 billion per year. For a discussion of the projections see Safadi and Laird (1996). For a survey of the
various estimates see Rodrik (1994). See also Martin and Winters eds (1996) and Srinivasan (1998).
7
In one study the gain to developing countries was estimated at US$90bn, or roughly one third of the
US$270bn total gain predicted at the time (OECD 1993).
many of the poorest countries in the world will actually be worse off
as a result of the round. Some estimates report that the 48 least
developed countries are actually losing a total of US$600 million a
year as a result of the Uruguay Round.8 While some least developed
countries will receive net gains, most will be net losers from the
round when all of the agreements are fully implemented. A large
share of the net losers are among the poorest countries in the world,
in particular sub-Saharan Africa, which has been estimated to have
lost US$1.2 billion as a result of the round (UNDP 1997:82).
One reason for this was that the modelled scenarios were not fully
reflected in actual agreements and the subsequent events.9 Several
reforms which were significant sources of predicted gains did not
proceed as had been hoped early in the negotiations. For example, the
Agreement on Textiles and Clothing (ATC) was structured to back-
load liberalization significantly;10 the ability of tariff-rate quotas
(TRQs) to liberalize agricultural market access was overestimated;
and the costs of implementation were almost completely ignored.
In addition the Uruguay Round agenda reflected, in large part, the
priorities of developed countries. Market access gains, for example,
were concentrated in areas of interest to developed countries and
there was only marginal progress on the priorities of developing
countries (particularly in agriculture and textiles). The result of this
regressive asymmetry was that after the implementation of
Uruguay Round commitments, the average OECD tariff on imports
from developing countries is four times higher than on imports origin-
ating in the OECD (Laird 2002). Domestic protection (particularly
agricultural subsidies) is also much higher in developed countries,
amounting to more than US$300 billion in 2002. The impact of this
protection is particularly regressive since producers in the poorest
developing countries are those most affected by OECD policies.
8
That is, US$600m in net losses. Some of the least developed countries gained from the round, but the
losers lost US$600m more than the sum of the gains.
9
The models themselves also make assumptions that may not be fully appropriate for less developed
countries. See Charlton and Stiglitz (2004).
10
As we noted, the developed countries were given a decade to remove their textiles quotas; the argu-
ment was that the extra time would allow them a smoother adjustment process. In practice, since little if any
adjustment has occurred, only the day of reckoning has been postponed. In the United States, the legisla-
tion implementing the Uruguay Round, by not adopting a steady phase-out of the quotas, made clear that
this postponement was the real motivation.
Only 4 per cent of the exports of developed countries are subsidized

by another WTO member, but 6.4 per cent of the exports of middle-
income countries are subsidized. By contrast, a much larger share
(29.4 per cent) of the exports of the poorest countries (not including
China and India) is subsidized by another WTO member.11
As well as receiving only a small share of the gains from the
Uruguay Round, developing countries accepted a remarkable range
of obligations and responsibilities. New trade rules and domestic
disciplines were introduced, but they too reflected the priorities and
needs of developed countries more than developing countries (e.g.
subsidies were permitted for agriculture, but not industrial prod-
ucts). Many of the rules acted to constrain the policy options (such as
industrial policies) of developing countries, in some cases prohibit-
ing the use of instruments that had been used by developed countries
at comparable stages of their development. Many of the new obliga-
tions imposed significant burdens on developing countries. In return
the least developed countries were promised financial assistance
with implementation costs and extensions of preferential market
access schemes. The common feature of these commitments is that
they were non-binding on developed countries. As a consequence
developing countries found themselves at the mercy of the goodwill
of developed countries. As Finger and Schuler (2000) aptly note: ‘the
developing countries took bound commitments to implement in
exchange for unbound commitments of assistance’. Insufficient
attention has subsequently been paid to the enormous demands
upon developing countries in implementing the outcome from the
Uruguay Round. Agreements related to intellectual property, cus-
toms valuation, technical barriers to trade, and agricultural food
safety have been particular targets of criticism in this regard.12
11
These figures may underestimate the relative effects of subsidies if developing countries’ exports are
more concentrated in those agricultural products which attract subsidies.
12
Many developing countries have been unable to meet their Uruguay Round obligations because of
these high costs. By January 2000, up to 90 of the WTO’s 109 developing country members were in viola-
tion of the SPS, customs valuation, and TRIPs agreements. Estimates of the cost of compliance to the
Uruguay agreements vary widely depending on the quality of the existing systems and the strength of insti-
tutions in each country. Hungary spent more than US$40 million to upgrade the level of sanitation of its
slaughterhouses alone. Mexico spent more than US$30 million to upgrade intellectual property laws. Finger
(2000) suggests that for many of the least developed countries in the WTO compliance with these agree-
ments is a less attractive investment than expenditure on basic development goals such as education.
Thus in the years after the Uruguay Round was completed, devel-
oping countries became increasingly disillusioned with its results.
There were continued tariff and non-tariff barriers by the rich
countries on developing-country exports and developing countries
had to implement their own obligations under the new agreements,
such as those on TRIPS, TRIMs, and agriculture. The promised bene-
fits from liberalization had not materialized. In 1999, in response to
concerns expressed by many developing countries that the Uruguay
Round agreements had produced an unbalanced result, the acting
European Trade Commissioner, Sir Leon Brittan, said: ‘the only way
to address the issue is through a new round of negotiations. Indeed, I
would ask all WTO members, including developing countries,
whether they are entirely happy with the present trading system. If
the answer is no, it is clear that the only way of improving upon that
system is in a new round.’13
As with previous rounds, the establishment of the Doha Round
was rooted in the experience of the past.
Unfinished business from Uruguay
The Uruguay Round left many loose ends. The 1994 Agreement on
Agriculture defined a framework in which agricultural protection
could be negotiated in the WTO, but it did not deliver significant
benefits to developing countries. Martin and Winters (1995) note that
the Agreement on Agriculture achieved ‘little in terms of immedi-
ate market opening’. The most significant element of the agreement
was the requirement that non-tariff barriers be converted into tariffs
that provide an equivalent level of protection. Since negotia-
tions over tariff reductions are easier than other forms of protection,
this ‘tariffication’ process was supposed to provide a simpler frame-
work for the long-term reform of agricultural trade in the future.
However, the Uruguay Round itself actually achieved very little in
13
Statement at the World Trade Organization High Level Symposium on Trade and Development, 17–18
Mar. 1999, reported in Sustainable Development 12: 2 (22 Mar. 1999). See also Raghavan (1999).
the way of liberalization. Although some reductions in tariff rates

were agreed, countries were able to use the tariffication process to
set very high initial tariffs, such that even after the agreed reform,
the new binding tariff rates were higher than the prevailing tariff
rates that had been in place for some years.
There was similarly little progress in the reduction of subsidies.
Indeed the overall level of OECD farm protection was not noticeably
reduced. In 1986–8 farm subsidies were equivalent to 51 per cent of
all OECD farm production, and fourteen years later, after the imple-
mentation of Uruguay commitments, at more than US$300 billion,
they still accounted for 48 per cent of all farm production (OECD
2003). Trade-distorting measures of industrialized nations displace
the agricultural exports of developing countries. By suppressing
world prices, these policies have a direct effect on farm incomes.14
Moreover there may be long-term effects, as investment is also sup-
pressed in countries whose trade is adversely affected by OECD
support.15
In non-agricultural goods, there is also scope for further liberaliza-
tion. The significant liberalization of manufacturing tariffs in devel-
oped countries over the last two decades might suggest that there is
little to gain from further negotiations on industrial products.
However, if this is true to some extent for developed countries, it is
certainly not the case for developing countries. While average devel-
oped country tariff rates are low, these nations maintain high barriers
to many of the goods exported most intensively by developing coun-
tries. When weighted by import volumes, developing countries face
average manufacturing tariffs of 3.4 per cent on their exports to devel-
oped countries, more than four times as high as the average rate faced
by goods from developed countries, 0.8 per cent (Hertel and Martin
2000). Moreover aggregate data hides the existence of tariff peaks. In
the United States, post Uruguay Round tariff rates on more than half
14
Estimates of the downward impact on world prices caused by OECD domestic support are between
3.5 and 5% for many agricultural commodities including wheat and other coarse grains and oilseeds
(Dimaranan, Hertel, and Keeney 2003).
15
Diao, Diaz-Bonilla, and Robinson (2003) report that protectionism and subsidies by industrialized
nations cost developing countries about US$24bn annually in lost agricultural and agro-industrial income.
Latin America and the Caribbean lose about $8.3bn in annual income from agriculture, Asia loses some
$6.6bn, and sub-Saharan Africa close to US$2bn. These estimates do not include dynamic effects.
of textile and clothing imports are between 15 and 35 per cent, while
in Japan 22 per cent of textile imports face tariffs of 10–15 per cent
(UNCTAD 1996).
Similarly in the processed food sector, Canadian, Japanese, and EU
tariffs on fully processed food are 42, 65, and 24 per cent respectively.
By contrast, the least processed products face tariffs of 3, 35, and 15 per
cent in these countries (World Bank 2002). Such tariff escalation
serves to discourage the development of food processing in least devel-
oped countries since the effective tariff rate on ‘value added’ in food
processing is very high. Tariff escalation and tariff peaks are mani-
festly unfair and have a particularly pernicious effect on development
by restricting industrial diversification in the poorest countries.
After the Uruguay Round, there was also a widely held view that
the TRIPS Agreement needed to be reviewed, particularly in its
application to public health and bio-piracy. Article 71.1 provided for
a review of the implementation of the TRIPS Agreement after year
2000, and for possible reviews ‘in the light of any relevant new devel-
opments which might warrant modification or amendment’. Many
developing countries felt that the Agreement as it stood primarily
reflected intellectual property rights protection suitable for devel-
oped countries, but largely disregarded important factors in develop-
ing countries.
International rules for intellectual property rights have poten-
tially huge public health effects and global distributional con-
sequences. Unbalanced rules—and there is a concern that present
rules are unbalanced—can impede efforts to close the North–South
‘knowledge gap’. Additionally the WTO also has the responsibility
to protect indigenous knowledge. While there have been a few dra-
matic bio-piracy cases,16 the full impact of expanded patentability
remains uncertain. Patent laws need to be changed so that the onus
of proof is reversed and companies should give an undertaking that
the patent they are seeking is not based on traditional wisdom.
Finally, the Uruguay Round imposed strong restrictions on devel-
oping countries’ use of industrial policies—policies that had arguably
16
In May 1995 the US Patent and Trademark Office (USPTO) granted to the University of Mississippi
Medical Center a patent (no. 5,401,504) for ‘Use of Turmeric in Wound Healing’. They revoked the patent
after dozens of references to the procedure were found in texts from India.
played an important role both in the development of Western

economies in an earlier century17 and more recently in the East Asian
Miracle.18 But they allowed developed countries to continue to use
non-tariff barriers to exclude goods from the developing countries.
Developing countries needed more freedom to use industrial policies,
and more protection from abuses by developed countries of dump-
ing duties, countervailing duties, safeguard measures, and phyto-
sanitary conditions.
New areas of importance
Services represent an increasingly large share of GDP and trade in

both developed and developing countries. With manufacturing
dwindling to 14 per cent of US GDP, it was natural for the US to shift
the focus of trade liberalization to services.19 Indeed, the irony was
that increasingly, it seemed as if the trade agreements of the past,
centered as they were around manufacturing, would, in the future,
be of greater benefit to China than to any other country.
But the Uruguay Round focused on the liberalization of those service
industries of primary interest to firms in OECD countries (like finan-
cial services). There was significantly less attention given to low-
skilled labor-intensive services in which developing countries have a
comparative advantage.20 Developing countries have increased their
exports of services more than fourfold since 1990, despite the large
trade barriers facing many of their most promising industries, such as
construction, shipping services, and health services (OECD 2002). In
these industries developing countries have legitimate and substantial
interest in the outcome of a new round of liberalization.21
17 18
For a discussion, see e.g. Chang (2002). See e.g. Stiglitz (1996).
19
Additionally, given the apparent barriers to service trade, there might be large gains from liberalization.
(See Brown, Deardorff and Stern (2002). They estimate that the global gains from service liberalization are
as high as $400bn. However, these estimates may overstate the benefits from liberalization if many of these
barriers are exogenous and not related to economic policy.)
20
Developing countries are capturing a growing share of trade in services. More than one quarter of the
world’s top 40 service exporters in 2002 were in developing countries.
21
As we discuss in the next section, these labor-intensive services are not the ones that have been given
priority in the Doha Round so far.
Some of the areas of service sector liberalization that were advanced

in the Uruguay Round may well have disadvantaged the developing
countries. Financial market liberalization, for instance, may have
weakened domestic financial firms, reducing the already scant supply
of credit available to domestic small and medium-sized enterprises.
This agenda of ‘new issues’ and ‘unfinished business’ is markedly
different from the agenda of the Doha Round in spite of the fact that
it was supposed to be a development round. The new issues which
had been put on the agenda at the Singapore Ministerial Meeting, the
so-called ‘Singapore Issues’,22 all centered around concerns of the
developed countries. There was one issue, competition policy, which
in principle could have been of benefit to the developing countries—
had dumping duties been brought into the discussion. But the devel-
oped countries were adamantly opposed to this.
A new sense of global responsibility

for development
On 30, November 1999, the World Trade Organization convened in

Seattle, for what was to be the launch of a new round of trade negoti-
ations. The negotiations were unsuccessful, but they were quickly
overshadowed by massive and controversial street protests. From
5 a.m. on the morning of the first day of the conference, several hun-
dred activists arrived in the empty streets near the convention cen-
ter and began to take control of intersections. As the day began, a
number of marches began to converge on the convention center from
different directions. A group of students marched from the north and
22
As noted earlier, these centered around (1) government procurement; (2) trade facilitation; (3) com-
petition; and (4) investment. The names, however, are somewhat misleading. ‘Competition’ did not focus,
for instance, on anti-trust matters. The developing countries had already expressed their hostility to the ini-
tiative by the OECD for a multilateral investment agreement. There was no reason to believe that the WTO
provided a venue in which an agreement acceptable to the developing countries could be worked out. In
any case, it was clear that this was an initiative of the developed, not of developing countries. Similarly, while
developed countries hoped to have greater access to government procurement in developing countries,
there was little hope that developing countries could make much inroad into procurement by developed
countries, especially in the central area of defense. This too was a developed-country agenda item.
were met by a march of citizens of the developing world who came

from the south. There was some violence, although the vast major-
ity of demonstrators were committed to peaceful forms of protest
including rallies, teach-ins, and street parties. The scale of the
demonstrations—even the lowest estimates claimed there were
more than 40,000 people—dwarfed any previous protest associated
with global issues of economics and equality.
The protests certainly did not cause the failure of the WTO nego-
tiations at the Seattle meeting. In particular, the negotiators from
developing countries were skeptical about the new issues being
placed on the agenda for the proposed round. They feared that any
compromise on their part on issues to be included in the negotiating
agenda would hurt them in the subsequent negotiations. However,
even if the effect of the protests on the negotiations is often over-
stated, the significance of the occasion as the coming-out of the anti-
globalization movement in the United States was profound.23 Over
the next two years the movement gained momentum with protests
in Melbourne,24 Prague,25 Washington,26 and Genoa.27 Suddenly the
gap between rich and poor was becoming shocking in the public
consciousness.
The global protest movement coincided with a significant increase
in the activism of civil society organizations and development-
focused NGOs such as the Catholic Agency for Overseas
Development (CAFOD), Christian Aid, and Oxfam. It also coincided
with a surge in international support for alternative development
initiatives and popular social movements, such as the Jubilee 2000
campaign to drop the debt burden of the developing countries and
the establishment of the World Social Forum. These groups have
exerted strong pressure on international political and economic
23
The anti-globalization movement developed in opposition to the perceived negative aspects of glob-
alization. The term ‘anti-globalization’ is in many ways a misnomer, since the group represents a wide range
of interests and issues and many of the people involved in the anti-globalization movement do support
closer ties between the various peoples and cultures of the world through, for example, aid, assistance for
refugees, and global environmental issues.
24
More than 5,000 people protested at the Asia-Pacific Summit of the World Economic Forum (Davos
Forum) in Melbourne in September 2000.
25
More than 10,000 people protested in Prague during the IMF, World Bank, and G8 Summit in
September 2000.
26
In April 2000, there were protests at the IMF and World Bank headquarters in Washington.
27
More than 100,000 people protested at the G8 meeting in Genoa in July 2001.
policy institutions to break the cycle of increasing world poverty,

inequality, and economic instability in developing countries.
At the same time developed countries made unprecedented com-
mitments within international institutions to bring development
issues to the center of international relations. At the United Nations
Millennium Summit in New York in September 2000, world leaders
placed poverty eradication at the heart of the global agenda by adopt-
ing the Millennium Development Goals (MDGs), which set clear
targets for reducing poverty, hunger, disease, illiteracy, environmental
degradation, and discrimination against women by 2015. Following
this, at the International Conference on Financing for Development,
held in Monterrey, Mexico, in March 2002, the advanced industrial
countries committed themselves to helping provide the finance for
development priorities. The Johannesburg Summit on Sustainable
Development in September 2002 attempted to establish an action
plan for the tasks to be completed in the coming years in order to
ensure sustainable global development.
Together these events signaled a new sense of collective respons-
ibility for the shocking poverty in developing countries and on
increasing recognition of the need for global collective action. The
addition of development concerns to the new round of WTO negotia-
tions was a natural part of a program of global collective action to
reduce poverty. It would have made no sense to exclude trade from the
poverty reduction initiatives because, as former US President Jimmy
Carter said during the Johannesburg Summit, ‘We cost the developing
world three times as much in trade source restrictions as all the over-
seas development assistance they receive from all sources.’
In this context, the Doha Ministerial Declaration adopted on
14 November 2001 contained, in its second article, an affirmation
of the new international commitment to development: ‘International
trade can play a major role in the promotion of economic develop-
ment and the alleviation of poverty. We recognize the need for all our
peoples to benefit from the increased opportunities and welfare gains
that the multilateral trading system generates. The majority of WTO
members are developing countries. We seek to place their needs and
interests at the heart of the Work Programme adopted in this
Declaration.’
It is certainly true that this new collective responsibility for the

problems of developing countries is in its infancy and has been char-
acterized so far by promises more than results. It is also true that the
issues of international equity embodied in the agreements at the
Millennium Summit, and followed up at Doha, Monterrey, and
Johannesburg, have not been precisely defined, but that does not
mean that they are not relevant. For a long time, arguments about
equity have held sway in domestic debates, and they should be wel-
comed into the international arena. Just as the distributional impacts
of domestic programs have become an important force in shaping
legislation in the democracies of the advanced industrial countries,
so too can development and distributional concerns become an
important force in the outcome of international bargaining.
4
What has Doha Achieved?
3
US Trade Representative Robert Zoellick’s statements after the
Doha Ministerial Conference reflected the prevailing positive mood
and the hope for a successful round of negotiations:
Doha lays the groundwork for a trade liberalization agenda that will be a
starting point for greater development, growth, opportunity and openness
around the world . . . we’ve settled on a program that lays out ambitious
objectives for future negotiations on the liberalization of the agriculture
market. These objectives represent a cornerstone of our market access pri-
orities for trade and they will create a framework that will help the United
States and others to advance a fundamental agricultural reform agenda.
On a range of issues, such as agricultural liberalization and reduction of
tariffs on non-agricultural goods, we’ve shown how our interests can con-
verge with the developing world. I believe that we in the United States
have an enhanced appreciation for the interests of developing nations
in trade.1
Despite the expressions of goodwill at Doha, progress on the

Development Round has been slow. Part of the problem is that,
while the interests of different developing countries differ, the evolv-
ing agenda itself was not really designed to reflect the real concerns
of developing countries. Throughout 2002 and 2003 it became appar-
ent that many developing countries felt that the Doha Round was
moving in the wrong direction on many key issues. They felt that the
new round offered them few immediate benefits but carried the risk
1
14 Nov. 2001, Office of the US Trade Representative. Online speech available at www.ustr.gov
of additional obligations. As a consequence developing countries

walked away from the Cancún Ministerial Meeting in September 2003.
Up to that point, Doha had achieved little progress on most of the
critical development issues. One of the key disappointments has
been agricultural reform, which many developing countries2 and
NGOs3 viewed as the primary objective of the round. The March
2003 deadline for agreement on agricultural modalities was missed.
When the US and EU finally presented a joint paper on agricultural
issues in August, the framework was widely criticized by developing
countries, correctly in our judgment, for ignoring their interests.4 On
the key issues of market access, domestic support, and export subsid-
ies the text was perceived to fall short of the level of ambition of the
Doha mandate; indeed, in some respects, what was offered was a step
backward. On domestic support, no specific figures were given for
reducing the most trade-distorting support. The text potentially
widened the scope for the use of production based financial support
(the so-called “Blue Box” support)—a step backwards in terms of lib-
eralization. Also the text did not focus on trade-distorting elements
of the “Green Box” measures (permissible forms of subsidy under
WTO rules). The framework was critically viewed by the Kenyan
Ambassador Ms Amina Chawahir Mohamed, who said that ‘the
EU–US text falls short of our expectations and as such we find it dif-
ficult to accept it as a basis of our further work’.
At the same time, agricultural initiatives within OECD countries
seemed to be undermining multilateral efforts. The US Farm Bill in
2002 increased the level of support to US farmers5 and strengthened
2
Sec. 7 of 6 June 2003 Communication from Argentina, Bolivia, Botswana, Brazil, Chile, China,
Colombia, Cuba, Dominican Republic, Ecuador, El Salvador, Gabon, Guatemala, Honduras, India, Malaysia,
Mexico, Morocco, Nicaragua, Pakistan, Paraguay, Peru, Thailand, Uruguay, Venezuela and Zimbabwe
(TN/C/W/13), makes it clear that ‘Reform of agricultural trade is of central importance for many develop-
ing countries’ and is ‘an essential ingredient of the negotiation and its outcome’ (emphasis in original).
3
Oxfam (2003) argues that ‘agriculture is the key to unlocking the Doha development agenda, and with-
out constructive steps on this issue, the broader negotiations cannot really restart’.
4
See the statements by Indian Ambassador K. M. Chandrasekhar, Brazil’s Ambassador Luis Felipe de
Seixas Correa, and China’s Ambassador Sun Zhenyu (TWN 2003).
5
The US Farm Security and Rural Investment Act (FSRIA) of May 2002 has a value of about US$190bn
over the next 10 years, about US$83bn more than under previous programs. It sets target prices which are
lower than the pre-1996 levels, but the total effective support is larger because average world commodity
prices have declined and the range of commodities included in FSIRA is larger than in the 1996 FAIR Act.
That act was intended to phase out farm subsidies, but even before the passage of FSRIA, farmers had
achieved additional support through emergency measures.
WHAT HAS DOHA ACHIEVED? 59
the link between subsidies and production decisions.6 One year

later, the EU’s 2003 Luxembourg reform of the common agricultural
policy (CAP) was also disappointing. The EU reform shifts support
from production-limiting subsidies (the so-called Blue Box subsidies)
to other, more acceptable forms of farm support (i.e. the Green Box
subsidies, which are deemed to be less trade-distorting). However, the
level of producer support will remain virtually constant—projected
to fall only from 57 per cent to 56 per cent (OECD 2004). Moreover
the reform has little impact on export subsidies or import barriers.
Both of these initiatives fell far short of expectations and signaled the
limited commitment of the US and EU to agricultural reform.
Consequently both plans had a depressing effect on the mood of mul-
tilateral agricultural negotiations.
After the Uruguay Round, there was a clear understanding that
there would be further liberalization of agriculture. There is now a
strong sense that the United States has reneged on that commit-
ment: whether the huge increase in agricultural subsidies is an
explicit violation of earlier agreements is of less importance than
that it represents a violation of the spirit of the agreement (or at least
what was taken as the spirit of the agreement by the developing
countries.)7 Just as the agreement has to be viewed as a whole, so too,
a Development Round agreement has to be viewed in the context of
the unbalanced agreements that preceded it.
In addition to their disappointment on agriculture, developing
countries are skeptical about the effects of the new items on the
agenda. There is significant opposition from developing countries to
the Singapore Issues.8 In the space of a month from early June 2003,
77 developing countries, including over half the WTO membership,
6
It provides counter-cyclical payments (CCPs) to US farmers which respond negatively to world prices.
This type of measure has allowed the US to dump its farm surplus on world markets. For example, the US
exports corn at prices 20% below the cost of production, and wheat at 46% below cost. See Cassel (2002).
7
The recent preliminary WTO ruling against American cotton subsidies (based on a complaint from
Brazil) has lent support to the critics. America claimed, remarkably, that their subsidies did not adversely
affect other cotton-exporting countries. Such claims clearly undermine the credibility of the position of the
developed countries.
8
Ministers from WTO member countries decided at the 1996 Singapore Ministerial Conference to
establish three new working groups: on trade and investment, on competition policy, and on transparency
in government procurement. They also instructed the WTO Goods Council to look into ways of simplifying
trade procedures, an issue sometimes known as ‘trade facilitation’. Because these issues were introduced
to the agenda at the Singapore Ministerial Meeting, they are often called the ‘Singapore Issues’.
made public statements urging that the Singapore Issues not be

included as part of the Doha Round.9 Since these issues are not
priorities for developing countries, their emerging centrality in
the agenda prior to Cancún was an incongruous feature of the
Development Round.
Several developing countries see the Singapore Issues as incur-
sions into their national sovereignty that are not justified by the ben-
efits they bring. Multilateral regulatory disciplines hold the specter
of repeating the worst elements of Uruguay by restricting the
options for individual governments to pursue development policies
based on their own national priorities and problems.
In addition there are concerns that the initiatives based on the
Singapore Issues may impose a large burden on the administrative
capacity of developing countries. There are significant costs associ-
ated with both the creation and enforcement of new regimes in com-
petition policy, investment regulations, and trade and customs
procedures.10 Many developing countries have been unable to meet
their Uruguay Round obligations because of these high costs.11
Another area where achievements have lagged behind rhetoric is
in the delivery of non-reciprocal trade preferences.12 These prefer-
ences are special market-access rights provided by developed coun-
tries to the poorest developing countries, i.e. they normally involve
a full or partial reduction of tariff rates on goods imported from very
poor countries. Recently there have been a number of initiatives in
OECD countries to discriminate further in favour of least developed
countries (LDCs). Most notable among these are the EU’s Everything
but Arms (EBA) initiative and the US’s African Growth and
Opportunity Act (AGOA), both of which remove tariffs on a wide
range of goods imported from the poorest nations. The EU has argued
that the EBA will ‘significantly enhance export opportunities and
hence potential income and growth’ for LDCs (CEC 2002). However,
9
CAFOD (2003) ‘Singapore Issues in the WTO: What do developing countries say?’.
10
Finger (2000) estimated the implementation costs of three of the Uruguay Round’s six agreements
that required regulatory change (customs reform, intellectual property rights, and sanitary and phytosanitary
(SPS) measures). His analysis suggests that the average cost of restructuring domestic regulations in the 12
developing countries considered could be as much as US$150m.
11
By January 2000, up to 90 of the WTO’s 109 developing country members were in violation of the SPS,
12
customs valuation, and TRIPS agreements. These preferences are discussed in Ch. 9.
studies show that preferential schemes have only limited impact on

LDC exports. Brenton (2003) concludes that trade in goods given
preferences for the first time under the EBA in 2001 amounted to just
two hundredths of one per cent of LDC exports to the EU in 2001.13
Even earlier preferences were not focused on goods exported by
LDCs: up to 50 per cent of the exports of non-ACP countries to the
EU did not receive preferential access and paid the most-favoured
nation (MFN) tariff (Brenton 2003). Overall, the impact of these
schemes has not yet been very significant, with the exception of
African apparel exports to the US under AGOA (World Bank 2003).14
The failure of the round to deliver on the expectations created at
Doha had, by the middle of 2003, sapped the goodwill that had been
created two years before. In particular, two issues had generated acri-
monious debate and attracted considerable adverse public attention.
Soon after the completion of the Uruguay Round, developing coun-
tries began seeking revisions of the TRIPS agreement. The agree-
ment required all member states to guarantee strong protection to all
intellectual property—rules which were widely thought to inhibit
technology transfer and exacerbate the technology gap between rich
and poor countries. In particular, the implications of the TRIPS
Agreement for developing countries’ access to life-saving drugs
attracted considerable attention. TRIPS allowed developing coun-
tries to acquire licenses enabling them to produce drugs domestically
at a fraction of the cost of the purchase price from patent-holding
corporations. But the licensing provision was useless for developing
countries with little or no manufacturing capability, since TRIPS
restricted trade in generic drugs between developing countries.
13
Part of the reason the EBA has had such a limited effect is that almost all EU imports from LDCs (more
than 99% in 2001) were already eligible for preferences under other schemes (Brenton 2003). In 2001
the EBA initiative granted duty-free access to imports of all products from the least developed countries
(except arms and munitions). Total exports from these LDCs to the EU increased by 9.6% in 2001.
However, in practice, the EBA was only relevant to the remaining 919 products (of the EU’s 10,200
tariff lines) which had not previously been granted duty-free status under either the GSP or Cotonou
Agreement. Of these 919 products, imports from LDCs were recorded in just 80 products in 2001. Brenton
(2003) notes that total exports of these products actually fell from 3.5m in 2000 to 2.9m in 2001.
Moreover trade in these goods in 2001 amounted to just two hundredths of one per cent of the total value
of LDC exports to the EU. Thus it appears that the direct impact of the EBA initiative has not been significant
in the short term and, given the small size of trade in affected products, is not likely to be large in the
medium term.
14
Moreover, beginning in 2008, cotton textile producers will have to use American cotton, further limit-
ing the benefits of AGOA.
Public support rallied around the plight of these countries, particu-

larly those in Africa dealing with the AIDS pandemic. The WTO and
the US (who had refused to consider revisions of the rules) attracted
public criticism which soured the atmosphere of the negotiations.
In August 2003, an agreement was finally reached when the US
changed its position to allow least developed countries to import
generic drugs from low-cost non-patent-holding producers in devel-
oping countries. But by 2003, the issue had done considerable dam-
age to the reputation of the WTO among developing countries and
their allies.
Just as public health concerns rallied opposition to the WTO’s
intellectual property rules, the plight of West African cotton farmers
served to focus world attention on the injustice of the rich countries’
agricultural policies. In 2001, cotton production in Benin, Burkina
Faso, Chad, and Mali and accounted for 5 to 10 per cent of gross
domestic product and an average of about 30–40 per cent of overall
export revenues. More than ten million people in West and Central
Africa directly depend on cotton production. These countries com-
plained that their cotton industries were suffering from cotton subsid-
ies in the US (and to a lesser extent in the EU). In 2001, rich countries
provided subsidies to their farmers which amounted to six times the
amount of their development aid, respectively US$311 billion and
US$55 billion. President Blaise Compaore of Burkina Faso addressed
the Trade Negotiations Committee of the WTO in Geneva on 10 June
2003. He claimed that these subsidies have caused economic and
social crises in African cotton-producing countries: ‘As a conse-
quence of cotton subsidies, in 2001, Burkina lost 1 percent of its GDP
and 12 percent of its export incomes, Mali 1.7 percent and 8 percent,
and Benin 1.4 percent and 9 percent respectively. The massive subsid-
ies awarded to cotton farmers in some WTO member states are
among the most important and direct causes of the problems encoun-
tered on the international cotton market. These subsidies artificially
inflate the offer and depress export prices.’ The African countries’
demands, including that the subsidies be reduced and compensation
be paid to African farmers, were swiftly rejected by the US.
In this climate of quashed expectations, WTO members met
for the Fifth WTO Ministerial Conference in Cancún, from 10 to
14 September 2003. The Doha Declaration had called for the Fifth
Ministerial Meeting to ‘take stock of progress in the negotiations,
provide any necessary political guidance, and take decisions as neces-
sary’ with a view to completing the round by January 2005. But, after
two years of missed deadlines and failed negotiations, there was
little progress to take stock of. In response to their frustrations,
developing countries united in several groups to strengthen their
negotiating voice. A group of twenty developing countries, the
G20,15 formed an effective negotiating block on agricultural issues.
Led by South Africa, Brazil, India, and China, the G20 pressed the US
and EU for greater market access and subsidy reduction. However,
the richest countries, as they had signaled on the cotton issue, were
unwilling to offer serious concessions: the developing countries
were frustrated again.
There was also disagreement in Cancún about the Singapore
Issues. As we have seen, these new issues in the round were primar-
ily of interest to developed countries. A diverse group of developing
countries, the G90, consisting of the least developed countries as
well as the other African, Caribbean, and Pacific countries, united in
their opposition to the inclusion of these issues in the round. When
negotiations began on the Singapore Issues on the morning of
Sunday, 15 September, the symbolism could not have been worse.
After several days of frustrating negotiations on the ‘development’
issues—during which the developing countries felt that the devel-
oped countries had not offered significant concessions—the agenda
turned to a set of issues that the developing countries feared would
restrict their policy freedom and impose large implementation costs
on them. At 3.30 on Sunday afternoon George Odour Ong’wen from
Kenya descended an escalator in Cancún’s convention center and
declared that Kenya had abandoned the negotiations. By 6 p.m.
Dr Luis Ernesto Derbez, the Mexican Finance Minister and chair of
the meeting, said that the positions of various countries were just
too diverse to make further debate worthwhile. The proceedings
were brought to a close and the meeting ended in failure.
15
The G20 consisted of Argentina, Bolivia, Brazil, Chile, China, Colombia, Costa Rica, Cuba, Ecuador, El
Salvador, Guatemala, India, Mexico, Pakistan, Paraguay, Peru, the Philippines, South Africa, Thailand, and
Venezuela. Membership however has been fluid.
Cancún represented a striking turnaround: less than two years

after the WTO’s Fourth Ministerial Meeting had ended in success at
Doha, the Fifth Ministerial Meeting had collapsed in bitterness.
Many developing countries refused to proceed with the agenda
because they came to the view that no agreement was better than a
bad agreement. The EU and the US blamed the developing countries
and asserted that they would be the main losers from the setback.
The EU Commissioner for Agriculture Franz Fischler dismissively
noted that if the G20 wanted ‘to do business, they should come back
to Mother Earth. If they choose to continue their space odyssey they
will not get to the stars, they will not get the moon; they will end up
with empty hands.’16
While some delegates berated the developing nations for their lack
of cooperation, others have argued that the assertion of power by
developing countries represented a victory for democracy. Developing
countries united to form voting blocs from which they could oppose
EU and US agricultural policies and make their case on new and
emerging issues. In addition to the G20 and G90 described above, the
G33 (the Alliance for Strategic Products and a Special Safeguard
Mechanism),17 led by the Philippines and Indonesia, wanted to main-
tain their ability to protect their own agricultural interests. Some
governments were supportive of these new groupings. The UK’s
Secretary of State for International Development, Hilary Benn, said
‘the better developing countries are able to articulate what they
want, the better chance we have in the end of reaching agreements
that will make a difference to poverty’.18
Little progress was made on any aspect of the negotiations in the
aftermath of Cancún. But by the middle of 2004, the EU and US top
trade negotiators (EU Trade Commissioner Pascal Lamy and US
Trade Representative Robert Zoellick) came back to the negotiating
16
Franz Fischler, ‘Ten Ingredients to Make Cancun [a] Success’, press conference, Brussels, 4 Sept. 2003.
17
The G33 actually comprises 42 developing countries of the WTO. They are: Antigua and Barbuda,
Barbados, Belize, Benin, Botswana, China, Cóte d’Ivoire, Congo, Cuba, Dominican Republic, Grenada,
Guyana, Haiti, Honduras, India, Indonesia, Jamaica, Kenya, Korea, Mauritius, Mongolia, Montserrat,
Mozambique, Nicaragua, Nigeria, Pakistan, Panama, The Philippines, Peru, Saint Kitts, Saint Lucia, Saint
Vincent and the Grenadines, Senegal, Sri Lanka, Suriname, Tanzania, Trinidad and Tobago, Turkey, Uganda,
Venezuela, Zambia, and Zimbabwe.
18
International Development Committee (IDC), Trade and Development at the WTO: Issues for
Cancun, Geneva, Seventh Report of Session 2002–03, HC400-I.
table with an attempt to salvage the round by offering a simplified

agenda and promises of compromise on some key issues.19 By August
2004, the General Council adopted a new framework for progress in
the round. The new agenda excluded all the Singapore Issues other
than trade facilitation. The round would focus on core market-
access issues in agriculture, services, and industrial goods. In particu-
lar the new text signals that the EU and US are willing to compromise
on agriculture and recommit themselves to special treatment for
developing countries.
In summary the agenda for the ‘Development Round’ has evolved
disappointingly for developing countries since Doha. It has done lit-
tle to address their concerns in agriculture and it has done little to
address problems posed by non-tariff barriers. It has not given prior-
ity to a developing-country service sector agenda and there have
been no reforms in basic procedures.
In addition, the proposed agenda’s new issues could have made life
worse for developing countries. The US wanted capital market liber-
alization as part of an investment agreement, even though the
weight of evidence was that capital market liberalization did not
promote growth but did lead to more instability. Under competition
policy, rather than creating a true competitive environment—
hindering the use of dumping duties as protectionist devices—there
was fear of restricting development and socially oriented preferences.
19
See the letter of 9 May 2004 by Pascal Lamy and Franz Fischler, and the letter of 11 Jan. 2004 to Trade
Ministers from Robert Zoellick. Letter to all Trade Ministers, unpublished.
5
Founding Principles: The Basis
of a Fair Agreement
3
One reason the Development Round is faltering is that the WTO

(like its predecessor the GATT) has been, by process and structure,
a mercantilist institution that has worked on a principle of self-
interested bargaining. The concept of a Development Round implies
a fundamental departure from the system of mercantilism towards
collectively agreed principles. However, there has been almost no
discussion, let alone agreement, on what such principles might be.
The lack of commonly agreed values has deprived the WTO’s mem-
bers of any means of collectively choosing a set of policies from
among competing proposals.
Progress in the Development Round needs to be accompanied by a
debate about principles, how those principles apply to trade, and
how they should be implemented in the current round of negotia-
tions. In this chapter we make a contribution to this debate by con-
sidering several fundamental questions: ‘What are the appropriate
boundaries for the WTO’s agenda?’, ‘What would constitute a ‘fair’
agreement?’, ‘What are the characteristics of a ‘fair’ negotiating
process?’ There are no universal answers to these questions, but
there are answers that derive legitimacy from commonly agreed
values implemented in a democratic process.
We begin with an analysis of the principles that should underlie

a development round of trade negotiations. It seems self-evident
that:
1. Any agreement should be assessed in terms of its impact on devel-

opment; items with a negative effect on development should not
be on the agenda.
2. Any agreement should be fair.
3. Any agreement should be arrived at fairly.
4. The agenda should be limited to trade-related and development-
friendly issues.
These principles may be widely agreed to; however, there may be

important differences in the way various terms such as ‘fairness’ are
interpreted and understood and about the meaning of terms and
about how to respond to conflicts among the principles.
Impact assessment
Any agreement should be carefully designed to promote, not hin-

der, development but there is surprisingly little economic analysis
of the precise consequences of various potential trade agreements
on participant countries. Where analytical studies have been done,
they have not penetrated to the core of negotiations and do not
seem to play a central role in setting the agenda. The absence of this
type of analysis raises the question of what is driving the prioriti-
zation of trade issues on the WTO agenda, other than a mélange
of prevailing orthodoxies and the momentum of special interest
groups.
The WTO Secretariat should be responsible for producing a
general equilibrium incidence analysis, analogous to what is con-
ducted when taxes are imposed, attempting to assess how different
countries are affected by different proposals. Publicly available
analysis would benefit developing countries, many of which are at an
information disadvantage relative to developed countries. Publicly
FOUNDING PRINCIPLES FOR A FAIR AGREEMENT 69
available information would also be an important source for

consumers, who are less equipped to lobby for favorable outcomes
than producer groups.
Analysis based on general equilibrium models must be sensitive
to the fact that different developing countries are likely to be affected
differently, and different groups within developing countries are
likely to be affected in different ways. For example, eliminating
developed country agricultural subsidies is likely to raise the price of
agricultural products, thereby benefiting countries that export these
commodities and hurting those that import them. Within individual
countries it is likely to benefit the producers of agricultural goods,
and hurt consumers. Thus the elimination of subsidies presents a
welfare trade-off for developing countries. But the net effect of the
elimination of subsidies is likely to be pro-development. Even if net-
importing countries experience aggregate losses, the reform has
potentially positive distributional consequences within the poor
countries since it is the producers (and those rural populations that
survive on agricultural production) who are among the poorest com-
munities in those countries. These communities are the most likely
to benefit, even if the country as a whole loses because it is a net
importer of subsidized commodities.
The results of general equilibrium models are sensitive to their
assumptions. Much of the analysis of the impacts (including, for
instance, judgments about whether particular types of agricultural
subsidies are trade-distorting) relies on a particular model of the
economy, the neo-classical model, which assumes full employment
of resources, perfect competition, perfect information, and well-
functioning markets, assumptions which are of questionable valid-
ity for any country, but which are particularly problematic for
Most of the tools used to analyse general equilibrium effects of trade
liberalization are static models. They describe the movement from
one ‘steady state’ to another but do not incorporate the costs associ-
ated with transition or the consequences for economies which are ini-
tially out of steady state. For example the models typically assume
that there is full employment. Trade liberalization measures are good
for a country because they enable resources to be redirected from
low-productivity protected sectors to more productive sectors as the

economy specializes in their areas of comparative advantage. Under
full employment, developing countries would unambiguously benefit
from trade liberalization measures, were it not for terms-of-trade
effects (changes in relative prices.) Most of the studies cited in
Appendix 1 that assess the impact on developing countries thus focus
principally on these terms-of-trade effects.
But with unemployment, trade liberalization is not needed to
‘release’ resources into more productive sectors and trade liberaliza-
tion may simply move workers from low-productivity protected sec-
tors into unemployment. This lowers the country’s national income
and increases poverty. There can be multiplier effects, so that the
total impact is far greater than the direct effect. Much of the opposi-
tion to trade liberalization arises because of the perceived effects on
employment. In more developed countries, monetary and fiscal pol-
icy should, in principle, enable the country to maintain nearly full
employment. As the advocates of trade liberalization repeatedly
emphasize, the objective of trade liberalization is not to create jobs,
but to increase standards of living by allowing countries to specialize
in areas of comparative advantage. But in many developing coun-
tries, with persistent unemployment—with unemployment rates
sometimes in excess of 20 per cent1—it is evident that monetary and
fiscal policies are unable to maintain the economy at full employ-
ment. While the standard neo-classical models typically deployed to
assess trade impacts do not identify the impact of trade liberalization
on the equilibrium level of unemployment2 (by assumption there is
none), even if trade liberalization had no impact on the equilibrium
level of unemployment, it may take the economy considerable time
to adjust, and the costs of adjustments—lost income and increased
poverty—may be considerable.
Another important assumption made in most of the analyses is
that there is no uncertainty, no risk. But changes in trade regimes
affect countries’ exposure to risk. In the absence of good insurance
1
In 2001 average unemployment rates reached 14.4% in Africa; 12.6% in transition economies; and
10% in Latin America. Such statistics, however, often under-represent the true level of unemployment—for
instance, the prevalent high levels of disguised unemployment.
2
See for instance the papers cited in Appendix 1, especially the contributions by Hertel (1997) and
Anderson, Dimaran, Francois et al. (2000).
markets, there can be first-order welfare effects arising from this

increased exposure to risk.3 For instance, with a quota, those who
compete with imports know precisely how much will be imported,
and therefore, if there is relatively little domestic volatility, they
will face relatively little price uncertainty. But with the tariffication
of quotas, countries are exposed to considerably greater volatility.4
It is important that any incidence analysis take into account other
pre-existing distortions. For instance, tax policies (often advocated
by international institutions), which effectively tax the informal
sector less than the formal sector, already distort production in favor
of the informal sector. In this context, trade regimes which lower the
international price of agricultural goods, typically produced by the
informal sector, have a larger adverse effect than would be the case if
tax policy were more neutral.
It is also important that any incidence analysis be based on an
assessment of the global general equilibrium impacts, which takes
into account the effects of a change in a trade regime on global rela-
tive prices. For instance, if a single small country were to subsidize
cotton, it would have a relatively small effect on the global price
of cotton. But if a large producer—e.g. the United States—subsidizes
cotton, it has an effect on the international price of cotton.
The fact that implementation and adjustment costs are likely to
be larger in developing countries, unemployment rates are likely to
be higher, safety nets weaker, and risk markets poorer are all features
of developing countries that have to be taken into account in con-
ducting a relative incidence analysis. If trade liberalization has a
large effect on inequality, then governments may be required to
strengthen their redistributive welfare system. Larger taxes generate
increased deadweight loss, which reduces the efficiency gains from
liberalization.
Large adjustment costs imply not only that the process of liberal-
ization should be conducted gradually, but also that there should not
be oscillations. Bilateral agreements on the way to a multilateral
3
For instance, Dasgupta and Stiglitz (1977) show that the change from quotas to tariffs may expose
countries to much greater risk. Newbery and Stiglitz (1984) show that the adverse effects from increased
exposure to risk may be so great that everyone in both countries may be worse off.
4
The incidence, in this case, depends on the extent to which there are disturbances in the domestic
markets, and the extent to which the external disturbances are correlated with the domestic disturbance.
agreement may be particularly detrimental when there are important

elements of trade diversion. The greater adjustment costs in devel-
oping countries may mean that the net benefit (net of the adjustment
costs into and out of the sector having temporary preferential treat-
ment) may be small and/or that there will, in fact, be relatively little
benefit, as expansion into that sector may be limited as investors
recognize its short-term nature.
Finally, it is of the first importance to distinguish between provi-
sions which should, in principle, make a country better off on its
own, almost regardless of the circumstances; provisions which
might or might not make a country better off on its own; and provi-
sions which essentially are redistributive in nature, with the gains
to one side being largely offset by losses to the other. We have argued
that many of the trade liberalization measures would, in a world of
full employment, make a country better off on its own. In the real
world, the question is often posed: ‘Why is there any need for a trade
agreement?’ Trade agreements are used only as a bargaining weapon.
The threat of not opening up one’s own market (which has a cost) is
used to force the opening up of the foreigner’s markets.
Note that many of the arguments that are currently used in favor
of certain provisions of proposed trade agreements contend that they
are good for the developing countries.5 To the extent that such argu-
ments are correct, of course, it implies that there is no need for a
trade agreement, other than the political economy argument that it
is only by bringing to bear the pressure from the gainers from trade
liberalization that one can overcome the resistance of the losers in a
world in which compensations are typically not made. Moreover, to
the extent that such arguments are correct, it implies that (apart
from global terms-of-trade effects) the issue of fairness only pertains
to the distribution of relative gains (relative costs and benefits) since
every country benefits. It also means that any country could unilat-
erally increase its gains simply by lowering its trade barriers further,
thereby expanding trade. To be sure, some of the opposition against
5
This includes not only trade liberalization, but also investor protections. Investor protections will attract
more investment. But if that is the case, then countries will have an incentive to undertake such actions on
their own. There are some legitimate worries, spelled out below: excessive investor protection may com-
promise general welfare concerns, e.g. about safety or the environment.
trade liberalization comes from those who would be hurt by it, in

particular special interests who benefit from protection; but some
opposition may arise because particular countries may be adversely
affected as a whole.
These cases need to be distinguished from intellectual property
rights, where stronger intellectual property protections may
increase the incomes of those in the more advanced industrial coun-
tries at the expense of those in the less advanced industrial coun-
tries. Here the issue is primarily redistributive, and is accordingly
fundamentally different from those that arise in connection with
trade liberalization.6 Again, the developed countries make a self-
serving argument that stronger intellectual property protections
will (1) induce more research or (2) induce more investment in
intellectual-property-intensive industries. But for most goods there
is relatively little evidence that the incremental profits generated in
developing countries have much impact on research. This is cer-
tainly the case for most drugs, where the overwhelming bulk of the
profits are generated by sales in the North—the drug companies do
little research related to illnesses whose primary incidence is in the
South. In many areas, such as soft drinks, trade secrets, not patents,
have been the basis of expansion into the South. In any case, again, if
it were true that stronger intellectual property protections led to
faster growth, countries interested in enhancing growth would
provide such stronger protection on their own.
Any agreement should be fair
Article 2 of the Doha Ministerial Declaration contains an implicit

principle of fairness: it focuses the Doha Round on the ‘alleviation of
poverty’ in developing countries and recognizes ‘the need for all our
peoples to benefit from the increased opportunities and welfare gains
that the multilateral trading system generates’.
6
See e.g. Bhagwati (2002), which argues, accordingly, that intellectual property should never have been
part of trade negotiations.
Similarly, the analysis in this book is based on a concern for social

justice and a degree of equity between nations. There are those that
would criticize this premise on the basis that notions of fairness
which are fairly well defined in contexts involving individuals are
not easily extended to bargaining among collectives such as nation
states. That brings us to the crucial question—should international
law, the rules and regulations that govern relations among countries
and among individuals and firms in different countries, be based
simply on realpolitik, or should considerations of social justice and
fairness play a role? This is a long-debated subject, and we cannot
pretend in this book even to touch on the merits of the alternative
positions. We believe that the view that international trade regula-
tions should be governed by principles rather than just economic
power is at least a respectable one—and one to which the interna-
tional community has seemingly subscribed. While we accept that
the application of principles of fairness is much weaker in the inter-
national context than it is inside nations, it is an important exercise
to explore the implications of principles of fairness between nations,
and that is what we do in this book.
Indeed, as globalization has proceeded, and there has been increas-
ing recognition of the need for global collective action, principles of
fairness have increasingly come to play an important role. The fact
that principles of equity underlying international relations are not
precisely defined does not mean that they are not relevant. Just as
arguments about equity hold sway in domestic debates, they are also
relevant in the international arena. International agreements require
the agreement not just of the rich countries, but also of the poor, and
if the citizens of democratic poorer countries believe that they are
being unfairly treated, they will refuse to agree or comply. Just as
information about the distributional impacts of domestic programs
has become an important force in shaping legislation in the democ-
racies of the advanced industrial countries, so too can information
about the development and distributional impacts of proposed inter-
national trade regimes become an important force in the outcome of
international bargaining.
Opponents of the view that fairness should be a consideration in
trade negotiations often argue that it is unnecessary because trade
agreements are voluntary, and so developing countries do not have to

sign up to any agreement which they think will make them worse
off. And since countries can withdraw from the WTO, the fact that
they do not do so shows that they must believe that they are net ben-
eficiaries. Such views almost certainly misconstrue the nature of the
power relationship between the developed and less developed coun-
tries. The advanced industrial countries are still able to get their
way, particularly by withholding aid unless developing countries
accept their demands. Moreover, countries may benefit from having
the ‘rule of law’ that the WTO provides for trade between nations—
the United States may, for instance, be restrained from the use of its
brute economic power but that does not mean that the rules them-
selves are fair, in any sense of the term. Still, the benefits of trade lib-
eralization may go disproportionately to the richer countries.
Moreover, those who put forth this argument fail to note the dif-
ference between individual and group actions. Given that other
developing countries had agreed to sign, it might pay any country
that is holding out to sign on; but it still may be true that the devel-
oping countries as a whole (or a subgroup of these countries) would
have been better off if they, as a group, had not signed. (The prison-
ers’ dilemma arises not only in the case of prisoners, but also in the
case of poor countries engaged in bargaining with the rich.)
Asymmetric trade agreements give rise to terms-of-trade effects—
changes in relative prices—which any single country may ignore,
but which become significant in a global agreement.7
However, although we believe that fairness is an important ele-
ment of trade negotiations, we do not pretend that it is easy to state
precisely its implications for the agreements. Underlying conflicts
about perceptions of fairness is the fact that because the circum-
stances of the different countries are different, any agreement that
applies ‘fairly’ or ‘uniformly’ to all countries may still have large dif-
ferential effects. This is why we have emphasized the importance of
an incidence analysis, an assessment of the differential effects on dif-
ferent countries. Any agreement that differentially hurts developing
7
The poorest developing countries play such a small role in international trade that even though their
tariffs may be high, their impact on global relative prices is small. This is not so for the advanced industrial
countries.
countries more or benefits the developed countries more (say, as

measured by the net gains as a percentage of GDP) should be pre-
sumptively viewed as unfair. Indeed, it should be essential that any
reform be progressive, i.e. that a larger share of the benefits accrue to
the poorer countries. This was almost certainly not true of the
Uruguay Round.
There is one key difficulty in interpreting this requirement. Many
of the costs of, say, agricultural subsidies are borne by the developed
countries. Not only are there huge budgetary costs associated with
the subsidies, but the subsidies distort production, and thus there is
an efficiency loss associated with them. Were developed countries to
eliminate their subsidies, they would therefore be among the main
beneficiaries. Thus, a refinement of the above criterion would look
at the benefits granted others: in competitive markets, it would be
reflected in the general equilibrium terms-of-trade effects received
by producers or paid by consumers; in non-competitive markets (or
markets with quota restrictions) it would be the value of access
granted.
One particular aspect of this should be emphasized: in trade dis-
putes, both de jure and de facto, the more developed countries are in
a better position to prevail. For instance, the costs to a developing
country of bringing a claim against a developed country or to defend
themselves against a claim from a developed country may be very
high; in practice the developing country is at a disadvantageous posi-
tion in any process entailing resort to complicated and expensive
legal proceedings.8
There is a long history within developed countries of those in posi-
tions of power using the legal system to maintain their privileges.
More recently, many developed countries have tried to come to terms
with the resulting inequities by providing public legal assistance.
Typically, because of the relatively low pay of those employed to
provide such assistance, this can go only a little way in redressing the
imbalance. But at a minimum, the developed countries should
8
More generally, as we discuss later, the WTO dispute system favors rich countries with the resources to
use it effectively for their own interests.
provide financial assistance to less developed countries involved in

such legal disputes to create a more level playing field.
By the same token, even were a developing country to prevail in a
WTO tribunal against the United States or Europe, the enforcement
system is asymmetric, and consequently unfair. The sanction for
violating a WTO agreement is the imposition of duties. If Ecuador,
say, were to impose duties on goods that it imports from the United
States, it would have a negligible effect on the American producer;
while if the United States were to impose a duty on goods produced
by Ecuador, the economic impact would more likely be devastating.
In practice, the WTO system has no effective way of enforcing penal-
ties against an unfair trade action whose main impact is on small
developing countries. When, of course, a major industrial country
takes a global action then there can be a global response (for example,
when the US increased protection on their steel industry, the whole
world responded and forced its removal).
But the other side of ‘fairness’ is the initial condition. Currently,
developing countries have higher tariffs than do developed coun-
tries.9 The United States might claim that it is only fair that devel-
oping countries cut their tariffs proportionately; this would entail a
greater amount of tariff reduction by the developing countries—and
accordingly the adjustment costs to the developing countries might
be greater.10 But the developing countries also point out that at the
very least the principle of progressivity should rule out adverse dis-
crimination against developing countries. Yet, currently, the devel-
oped countries impose higher tariffs against the developing
countries than they do against the developed countries, even taking
into account the so-called ‘preferences’.
Balancing these concerns are those dealing with historical
inequities. If a country’s relative weakness in part is due to a colonial
heritage, or more pertinently, to earlier unfair trade agreements (e.g.
that resulting from the Opium War in the nineteenth century in
China), to what extent does fairness and equity demand that current
9
For manufactured goods average tariff rates are 1.5% for developed countries and 11.5% for devel-
oping countries. For agriculture, average tariff rates are 15.6% for developed countries and 20.1% for
10
developing countries (Hertel and Martin 2000). See Ch. 8 for a discussion of adjustment costs.
agreements not reflect these past injustices? Trade negotiators from

the North would like to pretend that such inequities never occurred.
Those from the South might argue that one cannot separate events
today from the historical context.
The nature of trade agreements is, of course, that not every provi-
sion in the agreement is viewed to be ‘fair’. Some are intended to give
more to one party, others to another party; it is the package as a
whole which should be viewed as fair. But each trade agreement is
forward-looking: there are implicit and explicit understandings
about the direction of future agreements.
Fairness between foreign and domestic producers

While most of the discussion of this chapter concerns ‘fairness’
among countries, there is a related issue: fairness between domestic
and foreign producers. One of the purposes of trade liberalization is
to ensure that foreign producers are treated ‘fairly’. But again there
are questions: ‘What does fairness mean in this context?’ Foreign
producers and domestic producers are often inherently in different
situations. In the case of a developing country, the foreign producer
may have greater access to capital. He almost surely has greater
access to international technology. Much of the debate about protec-
tion concerns ‘leveling the playing field’, correcting these initial
inequities.
Most of the economics literature eschews the ‘fairness’ vocabu-
lary in favor of efficiency language. Protecting domestic firms is inef-
ficient: the country would be better off if it did not. But for reasons
hinted at in the previous section, these arguments are contentious.
There may be important learning benefits from protection. And
while economists have typically argued in favor of open subsidies
and/or government loan programs rather than the hidden subsidies
protection provides, direct subsidies may, for a variety of reasons, be
difficult or impossible to implement. As we discussed in Chapter 2,
in a second-best world, some protection may be efficient.11
11
For a historical argument, see Chang (2002). More recent theoretical analysis includes that of
Dasgupta and Stiglitz (1985).
Thus, there is contentiousness in both the efficiency and the

fairness arguments. But what cannot be justified in terms of either
are developed country non-tariff barriers, such as dumping, which
treat developing country producers disadvantageously relative to
their own, subjecting them, for instance, to a far higher standard for
what amounts to predatory behavior than that to which they subject
their own firms.12
By the same token, it is hard to justify demanding developing
countries to provide foreign firms with greater protections than pro-
vided to domestic firms. While there is some debate about the valid-
ity, or abuse, of the infant industry argument, there is no argument
for protection of the ‘grown-up’ industry’.
So too, America has found it desirable to impose lending require-
ments on its banks to ensure that they provide capital to underserved
communities, through the Community Reinvestment Act. Such
measures recognize that there is a role for government in encouraging
particular sectors of the economy. It seems unfair (and inefficient) to
preclude developing countries from undertaking similar measures.
Other problems in the interpretation of fairness

One of the most difficult issues is how to treat policy failures within
each of the countries. Suppose that it is true (as asserted earlier) that
the Uruguay Round in fact differentially benefited the United States.
But suppose the imbalance could have been reduced if only the
developing countries reformed their economies. They might, for
instance, have been able to benefit more from the reduction in tariffs
on manufacturing if only they had invested more in infrastructure,
so that they could have attracted more manufacturing.
By the same token, to what extent should the international trading
regime be blamed for inequities which arise, in part, because of how
other parts of the international system operate? Suppose, for
instance, that a ‘fair’ trade negotiation occurs within the WTO, but
12
For instance, the US anti-trust laws impose a very high standard for predatory pricing, much higher than
is employed in the US fair trade laws which pertain to the actions of foreign firms. Indeed, it has been argued
that using the US domestic standard, few foreign firms would ever be found guilty of dumping, but using
the dumping standard, most American firms could be found guilty.
that after the trade negotiations are over, the developing country has
to turn to the IMF for assistance, and that the IMF imposes further
trade liberalization as a condition for assistance. Viewing the two
negotiations together, as a package, clearly the developing country
may have given far more than it got within the trade package, but of
course it got, in addition, some foreign assistance. (Admittedly, in the
case of many of the bail-outs, the primary beneficiary of the bail-outs
may be banks in the more advanced industrial countries.) But even
apart from these demands that are put on developing countries, the
United States makes demands on other countries (Section 301 and
Super 301 actions13), to which they often feel compelled to accede.
Similarly, when international institutions encourage tax policies
which have the effect of distorting production towards the informal
sector, it implies, as noted above, that the West’s subsidies of agri-
culture have a greater adverse effect on the developing countries
than they otherwise would have had. In talking about the inequities
of the trade regime, should we assess its fairness coming on top of
distortions imposed or encouraged by the North, or in terms of what
the incidence would have been had a more neutral tax system been
imposed? Should we view the two actions together, assessing the
incidence of the two policies in conjunction, or should we only
assess the fairness of the trading regime itself?
By the same token, when countervailing duties are imposed
against a developing country which has ‘subsidized’ interest pay-
ments, by bringing them down from the usurious levels insisted
upon by the IMF to levels still slightly higher than in international
capital markets, is this unfair? Should the government only be
viewed as undoing a distortion? The problems are exacerbated by
demands (included in the recent bilateral trade agreements between
the United States and Chile and the US and Singapore) for capital
market liberalization. Capital market liberalization increases
economic volatility,14 and the increased economic volatility
increases the risk premium that investors demand,15 effectively
13
Super 301 authority—which expired in 1997 but was reinstituted in January 1999—enables the US
Trade Representative to identify the most significant unfair trade practices facing US exports and to focus US
resources on eliminating those practices.
14
See, for instance, Prasad, Rogoff, Wei et al. (2003) and Stiglitz (1999a, 2001, 2004).
15
See Stiglitz (2003).
increasing the interest rate charged. It seems unfair to force upon the
developing countries provisions which effectively increase the inter-
est rate they have to pay, and then, when the government tries to
undo the consequences, to slap a countervailing duty upon them.
In the South, there is a propensity to see such actions as coordin-
ated, pushed by special interests in the North. While they may see
more coordination than actually occurs, the impacts are often closely
akin to what they would be if they were coordinated. The high inter-
est rates, tax policies, and trade liberalization policies demanded by
the IMF do exacerbate the adverse effects on developing countries of
whatever trade liberalization measures they agree to within the
WTO. The two cannot be seen in isolation. This provides the basis of
one of the important recommendations that we make below.
With such disparate views of fairness, it is no wonder that the
South may feel that a trade agreement proposal is grossly unfair, and
yet the North might feel no pangs of conscience. Some might con-
clude that, as a result, we should simply drop the criterion of equity
as a desideratum of a Development Round agreement. That would be
a mistake. In a democracy, any trade agreement must be freely
entered into, and the citizens of the country must be persuaded that
the agreement is essentially fair. Moreover, there are several widely
accepted philosophical frameworks—in particular that of John
Rawls16—which at least provide some guidance for thinking about
whether any agreement is fair.
Any agreement should be fairly arrived at
Procedural fairness becomes an important complement to the kind

of fairness discussed in the preceding section (where fairness is
judged in terms of the outcomes) when there is some ambiguity
16
See Rawls (1971). Rawls’s method generates basic principles of justice that provide some guidance
in evaluating the fairness of particular trade agreement proposals—in particular, his ‘Difference Principle’,
that ‘Social and economic equalities are to be arranged so that they are . . . to the greatest benefit of the least
advantaged’ (Rawls 1971: 83). For alternative frameworks (which in the current context would arrive at
quite similar views) see Sen (1999).
about what should be meant by ‘outcome fairness’. Procedural

fairness focuses on the openness and transparency of the negotiation
process, and the manner in which the discussions are conducted. It
is hoped that the outcomes that are achieved through fair procedures
are more likely to be fair, though, of course, even an open and trans-
parent bargaining process is likely to result in ‘unfair’ outcomes
when the parties to the bargaining are of markedly different
strengths. But it should be clear: a fair agreement is unlikely to result
from an unfair process.17
Transparency is essential because it enables more voices to be
heard in the negotiating process and limits abuses by the powerful.
This is particularly important for developing countries, because of
the limited size of their negotiating teams. Of particular concern
is the lack of transparency of the negotiations. In the Uruguay Round
the developed countries negotiated via the infamous ‘green room’
methods, in which only a few chosen countries from the developing
world engaged directly in negotiations with the United States and
Europe. The ‘Green Room’ process has now been formally aban-
doned but the ongoing negotiating practice continues to place the
developing countries in a disadvantageous position because of the
complexity of the negotiations and their limited staffs.18 Procedural
fairness needs to deal with the asymmetry of power and the asym-
metry of information among WTO members. While the effects of
power disparities are difficult to reduce, informational disadvantage
can be remedied.19
The WTO’s Dispute Settlement System also lacks procedural fair-
ness in some important ways. In trade disputes, the system favors
developed countries both de jure and de facto. For example, the costs
17
There is now a large literature which establishes that setting the agenda may have a large effect on the
outcome; hence having a voice in the setting of the agenda is essential. The agenda in previous trade nego-
tiations has been unbalanced. This is evidenced by the fact that issues of benefit to the developed countries
have been at the center of the discussion. Issues like liberalization of unskilled labor-intensive services have
been off the agenda, while liberalization of skilled labor-intensive services have been on it. Since the bar-
gaining process affects the outcome of the bargain, the WTO needs to ensure that the process has clear
rules that ensure the effective participation of the weakest players.
18
See for example the open letter, dated 6 Nov. 1999, sent by 11 developing countries to the WTO chair-
man Ambassador Ali Mchumo of Tanzania, expressing their concern over the lack of transparency in the
WTO green room process.
19
Both increased transparency and the provision of (impact assessment) information, discussed in
Ch. 5, reduce information asymmetries.
to a developing country to attack a claim of intellectual property by

a Western company in a case involving bio-piracy may be very high.
In practice the developing country is at a disadvantageous position in
any process entailing resort to complicated and expensive legal pro-
ceedings. Thus the WTO dispute system favors rich countries with
the resources to use it effectively for their own interests. The EC,
Japan and the US were complainants in almost half (143 of 305) of all
bilateral disputes in the WTO Dispute Settlement system between
1995 and 2002. By contrast the 49 members classified by the UN as
Less Developed Countries did not bring a single challenge in that
period.20
To be sure, current arrangements were introduced because of
dissatisfaction with the previous GATT regime. The dispute settle-
ment system operated under the GATT was rarely used by develop-
ing country members, and thus the new system has been reformed
in an attempt to increase participation from all members. In some
areas there has been significant progress. The DSU has reduced the
length of proceedings, leading to more timely ‘trials’, introduced
automatic adoption of reports to remove the potential for a recalci-
trant defendant to block a ruling, and increased the consistency of
rulings through review by the appellate body. Taken together, it has
been predicted that these reforms would create a system in which,
unlike in the GATT years, ‘right perseveres over might’ (Lacarte-
Muro and Gappah 2000), and which would therefore entice develop-
ing countries to bring more cases to the WTO than they did to
the GATT.
However, the increasingly legalistic process has raised the trans-
action costs of settling disputes (Busch and Reinhardt 2003)21—one
factor that has contributed to the ongoing under-representation of
LDCs in the process. With the exception of Nigeria, which partici-
pated as a third party in the United States–Shrimp dispute, and the
Afro-Caribbean Pacific (ACP) countries, which participated as third
20
See Horn and Mavroidis (2003).
21
Welcome reforms have been implemented to prevent powerful defendants from delaying cases. In
particular, procedures and terms of reference have been standardized. One unforeseen consequence of
this is that it has increased the importance of the quality of the legal preparation before the case is heard
and, in some cases, increased the cost of initiating a case.
parties in the European Communities–Bananas dispute, no African

country and no other least developed country has participated in pro-
ceedings before the appellate body to date. This imbalance in partic-
ipation is also reflected in the outcomes of disputes. Whereas
developing country complainants have increased their success rate
under the WTO rules (the proportion of times defendants fully liber-
alize disputed measures)—up from 36 per cent of cases under the
GATT to 50 per cent under the WTO—developed countries have
been even more successful—their success rate rose by 40 per cent
under the GATT to 74 per cent under the new regime (Busch and
Reinhardt 2003).22
The low participation rates of developing countries and the imbal-
anced outcomes are partly a consequence of the asymmetry of the
final punishment mechanism. The final punishment mechanism is
triggered if the defending country continues to fail to comply with a
ruling after a ‘reasonable period of time’ for implementation.23 In this
event the complaining country may initiate action to seek authoriza-
tion to ‘suspend concessions or other obligations’. In practice this has
involved complainant governments imposing 100 per cent ad val-
orem tariffs on a list of products from the target country. However, in
the seven episodes where action was, authorized, the complaining
government actually performed the act in only three instances and in
each case the imposing government included a large country.24 This
22
Moreover 17 of the 24 WTO-era developing country complaints yielding full concessions came from
the wealthiest and most dominant developing countries (Argentina, Brazil, Chile, India, South Korea,
23
Mexico, Singapore, and Thailand). See Busch and Reinhardt (2003). DSU, art. 21.3.
24
This happened in Bananas, when the United States raised tariffs against the EC, and in Hormones,
when the United States and Canada raised tariffs on EC goods. For example, in Bananas, the United States
imposed 100% tariffs on the European Communities on a list of products that included, among others,
handbags and electrothermic coffee-and tea-makers. The Hormones case concerned an EC ban on imports
of beef from cows treated with hormones for growth-promotion purposes (oestradiol 17ß, progesterone
and testosterone, trenbolone acetate (TBA), zeranol, and melengestrol acetate). The EC claimed the ban
was necessary for food safety; the US and Canada claimed there was no evidence of harm to human health.
The WTO Panel found that the EC measure violated art. 3 of the Agreement on the Application of Sanitary
and Phytosanitary Measures: the EC measure was not based on these standards; it reflected a higher level
of protection and was not justified by a risk assessment, as required by art. 3.3. When the EC was unable to
implement the panel’s findings by the 13 May 1999 deadline, the US and Canada sought the right to retal-
iate to the amount of US$202m per year and CDN$75m per year. The arbitrators found the appropriate
level to be US$116m and CDN$11.3m per year, respectively. The United States raised tariffs against the EC
on a list of goods including Roquefort cheese and foie gras among others, and Canada’s list included
cucumbers and gherkins, among others. The four episodes so far where action has been granted but not
used are Bananas (Ecuador v. EC), Export Financing for Aircraft (Canada v. Brazil), Foreign Sales
Corporations (EC v. US), and Export Credits for Aircraft (Brazil v. Canada).
enforcement mechanism is a powerful weapon for large countries,

but a weak one for poor countries.
The policy space should be interpreted

conservatively
Defining the policy space appropriate for attention within the WTO
is a difficult task. There has been a tendency to expand the WTO’s
agenda to include all manner of international problems from intel-
lectual property rights to protection for foreign investors. The inter-
national community has found that bringing formerly intractable
international issues within the ambit of trade provides both a conveni-
ent negotiating forum and a ready mechanism for enforcement of
agreements. If the only test of inclusion in the agenda is that a policy
must affect trade flows, then the boundaries of WTO activity are
very hard to define because almost all international problems can
be linked to trade flows in some way. In this regard, policy-makers
have liberally employed the prefix ‘trade-related aspects of’ to
expand pragmatically the WTO’s mandate into a growing number of
issues.
However the growth of the WTO’s policy space comes at a price.
First, developing countries have limited capacity to analyse and
negotiate over a large range of issues. Second, the experience of the
Singapore Issues suggests that larger agendas burden the negotia-
tions. Third, the expansion creates room for developed countries to
use their superior bargaining power in trade negotiations to exploit
developing countries over a larger range of issues. For instance, when
the agenda was extended to competition policy, the issues relevant
to the foreign business interests of developed countries became the
main focus of negotiations while insufficient attention was given to
key areas of concern for developing countries, such as rules against
predation and the development of global anti-trust enforcement.
Similarly the focus of intellectual property negotiations has been
determined by the pharmaceutical industry in the industrialized
world. Almost inevitably, the determination of these issues will

reflect the consequences of the exercise of power.
For these reasons a ‘principle of conservatism’ needs to be intro-
duced to guide the growth of the WTO’s mandate. Further issues
should only be included in the agenda of a Development Round if
they score highly on three criteria: (1) the relevance of the issue to
trade flows, (2) its development-friendliness, and (3) the existence of
a rationale for collective action.
This third element reflects a general presumption in favor of
national sovereignty. There is no reason to force nations to under-
take certain actions unless their actions have effects on the trade of
others which require collective action to resolve. There are areas in
which a trade agreement is absolutely essential. These include an
international rule of law (procedures) for dealing with trade disputes
and/or agreements to prevent beggar-thy-neighbor trade policies.
There are areas in which international agreements would be benefi-
cial in managing cross-border externalities or global public goods.25
But modern trade agreements have been extended into areas which
intrude into national sovereignty with no justification based on the
need for collective action and without clearly identified and fairly
distributed global benefits.26 The presumption of consumer sover-
eignty is based on the premise that society should only interfere
with individual choices when those choices have consequences for
others, when there is a need for collective action, and the same is
true in trade.
25
For a discussion of the concept of global public goods, see Kaul et al. (2003). See also Stiglitz (1994,
1995).
26
Trade agreements might also be useful as a mechanism for governments to overcome domestic
political opposition to trade reform.
6
Special Treatment for
Developing Countries
3
There is considerable dissatisfaction with the treatment given by the
World Trade Organization to its poorest members. The development
of an appropriate framework which maintains the ‘rules-based’ trad-
ing system but differentiates between rich and poor countries is one
of the most important issues facing the Doha Round.
WTO members include 32 of the 50 least developed countries
(LDCs) recognized by the UN.1 The economic and social develop-
ment of these countries, in particular the eradication of extreme
poverty, is a major challenge for LDCs themselves, as well as for the
international community. Least developed countries are character-
ized by their exposure to a series of vulnerabilities and constraints,
such as limited human capital and productive capacity; weak insti-
tutions; geographical handicaps including poor soils, vulnerability
to natural disasters, and communicable diseases; poorly diversified
industries and underdeveloped markets for many goods and services;
limited access to education, health, and other social services; poor
infrastructure; and lack of access to information and communica-
tion technologies.
1
These are Angola, Bangladesh, Benin, Burkina Faso, Burundi, Cambodia, Central African Republic,
Chad, Democratic Republic of the Congo, Djibouti, Gambia, Guinea, Guinea Bissau, Haiti, Lesotho,
Madagascar, Malawi, Maldives, Mali, Mauritania, Mozambique, Myanmar, Nepal, Niger, Rwanda, Senegal,
Sierra Leone, Solomon Islands, Tanzania, Togo, Uganda, and Zambia. Eight additional LDCs are in the
process of accession to the WTO: Bhutan, Cape Verde, Ethiopia, Laos, Samoa, Sudan, Vanuatu, and Yemen.
Furthermore, Equatorial Guinea and Sao Tome and Principe are WTO Observers.
The WTO recognizes that the trade policies which maximize

welfare in the rich industrialized countries may not be the same as
those which do the most to promote development in the poorest
countries. The WTO recognizes the particular needs of developing
countries2 via ‘special and differential treatment’ (SDT)—special
provisions which give developing countries special rights and which
allow developed countries to treat developing countries more favor-
ably than other WTO members. Common SDT provisions include
support to help developing countries pay the costs of effectively par-
ticipating in the WTO, as well as aid to increase their capacity to
take advantage of new trading opportunities; exemptions from
agreements, which allow developing countries to choose whether or
not to implement agreements requiring regulatory or administrative
reform; and provisions allowing developed countries to give prefer-
ential market access to developing country members. In market
access, SDT gives developing countries greater freedom to use indus-
trial policies, including subsidies, and more latitude in tariff reduc-
tion (either higher bound tariffs or longer transition periods).
SDT is controversial. Proponents argue that the special circum-
stances of developing countries demand that they be given the
freedom to pursue industrial development through trade policies,
even if these policies involve some negative externalities for other
countries. Opponents express hostility to SDT because they see it
as an abrogation of the principle of reciprocity, and others believe
that SDT leads to protectionist trade policies which are inefficient
tools for industrial development and are likely to create vested
interests and misallocate resources. They argue that by granting
exemptions to WTO disciplines, members are doing developing
countries a disservice by encouraging protection, prolonging their
exclusion from the global economy, and denying access to the
benefits of openness.3 The neo-liberal view, embodied in the
2
There are no WTO definitions of ‘developed’ or ‘developing’ countries. Developing countries in the
WTO are designated on the basis of self-selection, although this is not necessarily automatically accepted in
all WTO bodies.
3
It should be noted that the benefits of openness are controversial. In particular, the relationship
between openness and growth is contentious. Rodriguez and Rodrik (2000) analyse the seminal paper by
Sachs and Warner (1995) which linked openness and growth. Rodriguez and Rodrik disaggregate the
Sachs–Warner openness measure and find that only the component relating to the black-market premium
is significant in growth regressions. They conclude that this indicates that macroeconomic stability, rather
SPECIAL TREATMENT FOR DEVELOPING COUNTRIES 89
Washington Consensus, is that most WTO rules, which commit

members to liberalize their trade policies, are good rules and should
be followed by poor countries as well as rich ones, i.e. the best
strategy for all countries and in all situations is to liberalize. The
neo-liberal approach relies on strong theoretical assumptions
about the efficiency and completeness of markets which, as we
discussed in Chapter 2, are unlikely to hold in developed countries,
let alone developing countries.
The alternative view places less confidence in markets and recog-
nizes a stronger role for government in economic development.
Markets are certainly powerful forces but where they are imperfect,
especially in developing countries, government intervention may be
required to correct their failures and make them work efficiently.
This view has been greatly influenced by the success of the East
Asian Tiger economies,4 whose reliance on market forces did not
preclude an active role for the government, including intervention-
ist policies in trade, FDI, technology transfer, and domestic resource
allocation (Lall 1992).
As noted earlier, neo-liberals responded to the East Asian success
by arguing that it was due to free trade and other non-interventionist
policies. To the extent that East Asian governments had been inter-
ventionist, they argued that this was a redundant and unnecessary
policy approach that coincided with the liberalization policies
which were in fact the true cause of their success (see World Bank
1993a and Noland and Pack 2003). Given that the issue remains
undecided, developing countries themselves should decide whether
they wish to use industrial policies which, it should be pointed out,
almost all of the successful industrialized countries used at similar
stages of development (Chang 2002).
At the same time, it is increasingly recognized that laissez-faire
policies may not be optimal in developing countries which suffer
from various types of market failures (see Chapter 2). Many devel-
oped countries currently have comparative advantage in agricultural
commodities and natural resources. However, reliance on production
than trade protection, matters for growth. Further analysis (e.g. Warner 2002) has revisited the issue, and
the relationship remains controversial.
4
See e.g. Stiglitz (1996).
of these goods can be problematic because historically, their prices

have been extremely volatile. In addition the production of these
goods may not embody the management and organizational skills
necessary to build an industrial economy. If this is true, then by con-
centrating production in the goods in which they have static compar-
ative advantage, developing countries may be preventing themselves
from establishing the foundations of an industrial economy.
However, the existence of justifications for government action
does not necessarily suggest that trade policy is the appropriate
instrument for intervention. Hoekman, Michalopoulos, and Winters
(2003) argue that the use of traditional trade policy instruments are
likely to impose more costs than benefits on developing countries.
These authors are right to claim that the case for various trade
instruments of industrial policy is controversial, but it is wrong to
conclude that there is a strong case for restricting the policy options
of developing countries on these grounds. It requires us to believe
that developed countries in the WTO are better informed (and per-
sistently so) than developing countries. In reality SDT provides
developing countries with additional freedom to use a range of
industrial policies, which are in some cases used efficiently, and in
some cases are not. However, the ex post identification of failed
industrial policy experiments is not necessarily evidence of bad pol-
icies. Rodrik (2004) points out that in the case of subsidies for entre-
preneurship, even under the optimal incentive program, some of the
investments will turn out to be failures. Failed industrial policy
experiments are examples of the ‘self-discovery’ of comparative
advantage. The costs associated with these failures may be more
than offset by the gains from the successes. All that is required to jus-
tify industrial policies is that their expected value be positive. Indeed
if there are too few failures, this might be a sign that the industrial
policy is not aggressive enough.
In addition there is a difference between openness and liberaliza-
tion. Developing countries with large populations working in pro-
tected industries must be mindful of the costs of adjustment. Trade
liberalization creates adjustment costs as resources are moved from
one sector to another in the process of reform. When tariffs are
reduced, import-competing firms may reduce their production in
the face of new competition, causing some of their workers and

capital to lie idle for a period. The firm’s laid-off workers will incur
costs while searching for new jobs and may need to invest in retrain-
ing. Governments will be called upon to provide assistance to the
unemployed while also incurring costs associated with implement-
ing the new systems to manage reform. In developing countries
which are already characterized by high unemployment, weak insur-
ance markets, and low levels of social protection, adjustment costs
may have a particularly severe effect.
Discomfort with the consequences of the neo-liberal view of spe-
cial and differential treatment led the Zedillo Commission, in its
‘Report of the High-Level Panel on Financing for Development’ to
the Monterrey Conference on Financing for Development in 2002, to
assert the following on infant industry protection: ‘However mis-
guided the old model of blanket protection intended to nurture
import substitute industries, it would be a mistake to go to the other
extreme and deny developing countries the opportunity of actively
nurturing the development of an industrial sector’ (Zedillo 2001).
Similar support for SDT was expressed in article 44 of the Doha
Declaration, which affirmed ‘that provisions for special and differ-
ential treatment are an integral part of the WTO Agreements’. The
article provides that special and differential treatment provisions
shall be reviewed in the Doha Round ‘with a view to strengthening
them and making them more precise, effective and operational’.
SDT in the Doha Round: Exemption,

not exclusion
At the WTO’s Fifth Ministerial Conference in Cancún, in September

2003, the negotiations collapsed and the meeting ended in failure.
One reason for the deadlock was that developing countries were
worried about being forced into accepting obligations which would
hurt their industries or impose large implementation costs on them.
Many developing countries had felt disadvantaged by the last round,
the Uruguay Round, and they came to the view that no agreement
was better than another bad agreement. After Cancún, developing
countries stepped up their demands for special and differential treat-
ment as a prerequisite for progress in the round. For wary developing
countries, SDT was an insurance policy—it would give them the
flexibility to opt out of any agreement which proved to be too oner-
ous for them. A group of developing countries, the G33,5 united
behind the issue of SDT. As the round began to regain momentum in
2004, they renewed their calls for SDT to be given a higher degree of
clarity and specificity. In particular they wanted the right to identify
special products of interest to developing countries on which there
would be no tariff reduction commitment and no new tariff-rate
quota commitments.6
In May of 2004, EU Trade Commissioner Pascal Lamy attempted
to placate the developing countries and salvage the round by offering
a significant compromise on SDT. In a letter to trade ministers he
wrote: ‘on agriculture and [industrial goods], we propose that the
least developed countries and other weak or vulnerable developing
countries . . . should not have to open their markets beyond their
existing commitments, and should be able to benefit from increased
market access offered by both developed and advanced developing
countries. So in effect these countries should have the “Round for
Free”.’7
The danger of the blanket approach to SDT embodied in the
‘Round for Free’ approach is that it may act as a disincentive to the
participation of developing countries in the Round. If the least devel-
oped countries are required to do nothing, they may be pushed to the
periphery of the negotiations. It would be unfortunate if the ‘Round
for Free’ rhetoric came to imply an agreement between developed
5
For a list of the membership of the G33, see the Glossary.
6
See the statement issued by Indonesia on behalf of the G33 member countries in Geneva,
28 July 2004.
7
See the letter of 9 May 2004 by Pascal Lamy and EU Agriculture Commissioner Franz Fischler. After
attracting criticism from other developed countries for his ‘capitulation’, Lamy quickly backed away from this
grandiose offer. In June he noted that the ‘Round for Free’ slogan was perhaps a misnomer since develop-
ing countries would be required to make commitments on binding their tariffs in some areas, and partici-
pating in negotiations in trade facilitation. He coined the somewhat less catchy slogan ‘Round at a Modest
Price’. See his speech ‘Where Next for EU Trade Policy?’ to the Deutsche Gesellschaft für Auswärtige Politik,
Berlin, 11 June 2004.
countries with only a peripheral role for developing countries. Then

the Development Round would bear a striking resemblance to the
early rounds of trade negotiations, in which the GATT operated as a
club for the advancement of rich-country interests. In those early
rounds, developing countries were burdened with few obligations,
but they had only a weak voice in the negotiations and little power
with which to assert their interests. Developing countries benefited
from industrial country liberalization, thanks to the Most Favored
Nation (MFN) principle,8 but their peripheral role meant that they
could exert little pressure on the way that industrialized countries
liberalized. Thus liberalization of goods of interest to developed
countries proceeded swiftly, but goods of interest to developing
countries, especially labor-intensive goods, lagged behind and devel-
oping countries ultimately suffered. Some developed countries were
happy with this system because the small poor countries did not
have sufficiently attractive markets to bother with: the benefits of
market access were smaller than the costs of liberalizing their own
labor-intensive import-competing sectors. The ‘Round for Free’
approach smacks of the same two-tiered system which exempted
developing countries from commitments but excluded them from
the negotiations. However, as Keck and Low (2004) argue, ‘where
new policy areas or new rules are under negotiation, or consideration
for negotiation, the best interests of developing countries would be
served through engagement with respect to the substance of core
proposals’. Another problem with the ‘Round for Free’ approach
which concerns many negotiators from developed countries, is that
it allows the poorest countries to retain a veto over a Round in which
they are contributing very little. This creates a ‘hold-up’ problem
which does not benefit their interests in the system—it merely
encourages the OECD nations to liberalize further outside of the
WTO system, which takes us back to the problems of previous
rounds.
Moreover the ‘Round for Free’ approach may result in forgone
opportunity costs on developing countries by robbing them of the
8
MFN is enshrined in the first article of the GATT, which governs trade in goods. MFN is also a priority in
the GATS (art. 2) and the TRIPS Agreement (art. 4), although in each agreement the principle is handled
slightly differently.
benefits of liberalization of South–South trade. If developing

countries suffer from barriers in other developing countries, then
proposals in the Doha Round to allow minimal liberalization by
poor countries could backfire. Developing countries now account
for around one-third of global trade. Intra-developing country, or
South–South, merchandise trade has grown at twice the pace of
world trade over the past decade. Yet barriers to South–South trade
are high. For example, Latin American exporters of manufactures
face average tariffs in the rest of Latin America that are seven times
as high as tariffs in industrialized countries (ESCWA 2004). East
Asian exporters face tariffs in other East Asian countries that are
60 per cent higher than in rich nations. Indeed, developing countries
stand to benefit a great deal from improved market access to other
developing countries. The World Bank (2002b) estimates that devel-
oping countries stand to realize welfare gains of more than US$30
billion per year if other developing countries eliminate tariffs on
industrial goods and a further US$30 billion if they remove their
barriers to agricultural trade.
A Doha Round Market Access Proposal
In this section we describe a framework—the Doha Market Access

Proposal (MAP)—which maintains the ‘rules-based’ trading system,
but differentiates between rich and poor countries.9
The challenge is to design special and differential treatment which
gives developing countries flexibility to deal with their development
problems and minimizes adjustment and implementation costs,
without marginalizing their participation in the global trading sys-
tem or forgoing the gains from South–South liberalization. To
achieve this, all WTO members could commit themselves to pro-
viding free market access in all goods to all developing countries
poorer and smaller than themselves. Thus all developing countries
could expect free access to all markets with (1) a larger GDP and
9
This proposal is based on Charlton (2005).
(2) a larger GDP per capita. This special and differential treatment
provision would bind developing and developed countries alike. For
example, a middle-income country like Egypt, with GDP per capita
of US$1,390 and total GDP of US$82 billion, would receive free mar-
ket access to countries like the United States, but would be required
to give free market access to a country like Uganda (GDP per capita
of US$240 and GDP of US$6.2 billion).
The principle underlying this proposal is that all countries should
participate in an enforceable system of preferential market access in
which rights and obligations are distributed progressively according
to objective criteria. The proposal presented in this section repre-
sents one straightforward means of implementing this principle.
Additional provisions for specific sectors, alternative dimensions to
differentiate between countries, implementation periods, and vari-
ous other complexities that would undoubtedly be part of any
applied version of this proposal are left out of the exposition in this
section.
Advantages
This proposal has several advantages over alternative schemes:
1. It involves significant liberalization. Figure 6.1 plots the GDP

and GDP per capita of WTO members. The dotted lines illustrate the
implications of the proposal for Egypt, a country in the middle of the
distribution in both size and wealth. MAP would require Egypt, after
a negotiated implementation period, to provide free market access to
more than fifty developing countries to its south-west in Figure 6.1
(with total market size of US$500 billion). In return it would receive
free market access to more than twenty developed and upper-
middle-income countries to its north-east in the figure (with a total
market size of US$28 trillion).10
As with existing preferential schemes, the effect of MAP would
to some extent be limited by rules of origin. It is beyond the scope of
10
In this calculation the EC is treated as one member; this makes little difference to the market size
numbers.
US
EU
JAPAN
CHINA
INDIA
INDONESIA
EGYPT
GDP (US$, log)
ISRAEL
UGANDA
ANTIGUA
GDP per capita (US$, log correlation = 0.6)
Figure 6.1. WTO members’ GDP and GDP per capita
this discussion to describe options for rules of origin that might

be required to implement MAP, but it is worth noting that MAP
would significantly reduce the distortionary effect of rules of origin
on LDCs’ trade compared to current SDT approaches, since many of
the middle-income countries from which LDCs might import
intermediate inputs would also receive preferential access under
MAP to some of the rich countries to which the final goods are
exported. Thus, while not eliminating the problem of rules of origin
in SDT, MAP would reduce its effect in practice compared to the
status quo.
2. In particular, it involves significant South–South liberalization.

Another advantage of MAP is that it takes South–South liberaliza-
tion seriously. Many existing types of SDT (and several proposed
changes) do little to promote South–South trade, the liberalization
of which, we have noted, could bring large gains to developing
countries. Indeed most Doha Round estimates indicate that the

scope for welfare gains for developing countries is larger from the lib-
eralisation by other developing countries than from liberalisation by
developed countries (Francois, van Meijl, and van Tongeren 2004).
South–South liberalization has progressed slowly. Attempts at
preferential market access agreements have been made outside the
WTO under the auspices of the Global System of Trade Preferences
among Developing Countries (GSTP).11 Unfortunately, the GSTP
is based on reciprocity—one reason for the low participation of
least developed countries among its members—and it has struggled
to make significant progress.12 Bilateral and regional free trade agree-
ments (FTAs) between developing countries are increasing in
number, but it should not be assumed that South–South FTAs are
unequivocally good for development, since they discriminate
against third-party developing countries and the margin of discrimi-
nation is higher than is the case in North–North FTAs because
developing country MFN tariffs tend to be higher. Thus, there may
be a strong case for introducing a development dimension into
South–South agreements. There are schemes being considered by
some larger developing countries including India, China, and Brazil
which would give special access to the least developed countries.
While additional market access would be welcome, these schemes,
like the existing GSTP schemes operated by the advanced industrial
countries, would be a patchwork of discretionary and conditional
promises rather than clear legal rights enforceable within the WTO.
Within the WTO, developing countries have often been urged to
reduce their MFN tariffs on the grounds that this would lead to an
increase in South–South trade. MAP recognizes that, for this
purpose, liberalization need not occur on an MFN basis.
3. Obligations are distributed progressively. MAP is progressive in

the sense that it requires significant South–South liberalization
from middle-income countries and very little from the poorest and
most vulnerable countries. It requires the most liberalization from
11
The GSTP, established in 1988 and promoted by UNCTAD, provides trade preferences to developing
countries without extending them to developed countries.
12
The two previous GSTP rounds, in the past two decades, were not as successful as expected, due to the
economic situation of the poorest developing countries and the poor negotiating capacity of member states.
the countries in the north-east of Figure 6.1 (in particular the Quad
countries—Canada, the EU, Japan, the US) and less of those in the
south-west of the figure (mostly African LDCs). Under this scheme,
all but the very poorest countries do not get the ‘Round for Free’,
since all countries accept the obligation to provide market access to
other WTO members smaller and poorer than themselves. In return
the developing countries receive considerably more market access,
under well-defined commitments, than under existing preferential
schemes, which are discretionary schemes operated by industrial-
ized countries that are not subject to detailed WTO regulation
governing their implementation.
All developing country WTO members benefit from the scheme.13
Even the largest and richest developing countries receive free access to
markets, whose total size is more than seven times the size of the mar-
kets to which they must give free access. The median ratio of market
access rights to obligations under MAP is 303 : 1, i.e. the median
developing country receives access to markets 303 times the size of
the markets to which it must give free access. Alternatively, measured
by imports, the median developing country receives free market
access to countries whose total imports are 113 times the size of the
imports of the countries to which it is required to give access.
4. Countries can manage major import threats. The proposal

imposes no extra obligation on developing countries to open their
markets to larger or more developed economies. This gives develop-
ing countries the option to provide their key industries with some
protection from imports from economies with cost advantages
derived from either the scale of their economies (e.g. larger coun-
tries, particularly China) or technological advantages (more devel-
oped countries). The prevalence of import barriers in some sectors
indicates that developing countries wish to protect particular indus-
tries from import competition. This protection may form part of an
industrial strategy based on ‘infant industry’ protection,14 but is
more likely to originate in a desire to avoid adjustment costs, which
13
For purposes here, developing countries are defined as those that had a GDP per capita below
US$10,000 in 2003.
14
See the discussion of chapter 2 for the controversy over the impact of such protection.
could be particularly severe in developing countries characterized by

high unemployment, weak risk markets, and low levels of social
insurance.
As Hoekman et al. (2003) point out, protectionism can be self-
defeating for developing countries in a world where multinational
corporations have made production increasingly fragmented inter-
nationally. For many developing countries, ‘the only option to reach
the minimum scale required for sustained growth in output is integ-
ration with the rest of the world’ (Keck and Low 2004).
This proposal facilitates integration by providing all developing
countries with significantly increased market access to larger and
richer markets, while providing the option of protection from
imports from countries which are at later stages of development or
have scale advantages.
5. It is consistent with other MFN liberalization schemes. It is

important to point out that this proposal is not opposed to openness.
It does not involve any increases in existing MFN rates. Each coun-
try would continue to apply MFN rates uniformly to larger and more
developed countries.
In addition, this proposal is squarely in the realm of SDT. There is
still a role for the WTO to negotiate MFN tariffs, i.e. it complements
other proposed modalities for MFN tariff reduction, rather than
replacing them.
One concern with the proposal is that it may affect the bargaining
positions of developing countries in future rounds. One of the unfor-
tunate side-effects of existing preferential schemes is that they create
an inbuilt incentive for developing countries to block MFN liberal-
ization which would erode their preference margins. But this prob-
lem is much less severe for MAP than for existing preference schemes
because MAP is far less distortionary: large rich countries do not
give LDCs preferences that they do not also give to middle-income
countries. Thus MFN liberalization by developed countries does not
cause LDCs to lose out relative to middle-income countries.
6. It transforms discretionary preferential schemes into well-

defined obligations within the WTO. One of the main advantages of
this proposal over existing types of SDT is that it delivers clearly

defined and legally binding rights to developing countries in a way
that existing preferences do not. Many of these existing preference
schemes were originally spawned by part IV of the GATT, which
includes provisions on preferential treatment for developing coun-
tries. This exception was further expanded in 1979 in the decision
which has come to be known as ‘the Enabling Clause’. This consolid-
ated the concept of ‘differential and more favourable treatment’ for
developing countries as well as the principle of non-reciprocity in
trade negotiations.15
However, the problem with this (potentially) wide-ranging clause
is that it has never placed any formal obligations on developed coun-
tries. Instead, piecemeal preferential deals have been established
which cover a limited range of goods from a limited group of coun-
tries. These preferences, the most important of which are offered by
the Quad countries, often divert trade from other poor countries.
Another problem with preferential schemes is their uncertainty.
Keck and Low (2004) argue that SDT should enshrine ‘legal rights
and obligations’, whereas existing preferences have become merely
‘legally unenforceable statements of intent or best-endeavour under-
takings’. Preferences are not binding on the countries which grant
them, and can be altered to exclude certain products or withdrawn
entirely at the discretion of the preference provider. For example,
in 1992 the US withdrew US$60 million worth of pharmaceutical
imports from their preference scheme because the US Trade
Representative determined that India had weak patent protection
which adversely affected US companies.
Without binding obligations, preference providers have faced
pressure from their own import-competing domestic lobbies to min-
imize the scope of their preferential schemes. As the Sutherland
report notes, it is ‘grantor, rather than grantee, country interests
[which] have determined the product coverage and the preference
margins in GSP schemes’ (Sutherland 2005).
15
The most significant provision of the Enabling Clause is that which enables members to accord
differential and more favorable treatment to developing countries as a departure from the MFN Clause.
It stipulates that ‘contracting parties may accord differential and more favourable treatment to developing
countries, without according such treatment to other contracting parties’.
7. It balances simplicity against the need to differentiate. The idea

that SDT should be provided to countries based on objective access
criteria has been previously addressed by Stevens (2002). Stevens’s
proposal suggests that a new SDT regime involve ‘greater differenti-
ation of treatment between WTO members which, in turn, implies
the establishment of objective criteria on which to determine the
differentiation’. For example, he suggests that access to some types
of SDT in agriculture should be based on measurable criteria relating
to food security, e.g. countries could qualify for special treatment if
they have a per capita calorie intake of less than a certain level (indi-
cating vulnerability to food insecurity) and a high share of agricul-
ture in GDP (indicating the importance of agriculture in livelihoods)
and a high share of food imports to GDP (indicating import depen-
dency). Stevens’s approach involves setting objective criteria on an
agreement-by-agreement basis. Thus, special and differential treat-
ment would be available to countries which met objective precondi-
tions indicating their need for exception and/or assistance. The
appeal of this approach is that it closely matches the needs of specific
countries to special treatment in different provisions. In addition, it
provides more certainty to developing countries since, once the con-
ditions are predetermined, eligibility would be automatic rather
than at the discretion of other WTO members.
The disadvantage of this approach is that it would add to the com-
plexity of trade negotiations and greatly increase transactions costs.
SDT measures are already overly complicated in many areas. Hudec
(1987) refers to preferences as systems of ‘refined complexity’,
determined by an ‘orgy of fine-tuning’. The process of tailoring
objective criteria for SDT in each agreement requires countries to
agree on measurable criteria and on eligibility cut-offs. As Stevens
(2002) himself notes, ‘the whole process is likely to be fraught with
political difficulty’. It is likely that neither the international con-
sensus on these issues nor the necessary negotiating capacity cur-
rently exists to operate such an ambitious and resource-intensive
SDT system.
By contrast, MAP is simple to negotiate. It would entirely do away
with the whole ‘spaghetti bowl’ of GSP preferences (although not
FTA preferences) and it would save the EU the bother of negotiating
the market access part of the economic partnership agreements with

African, Caribbean, and Pacific states.16
Moreover, it includes an inbuilt flexibility that removes the need
for renegotiation over time. As countries develop and overtake oth-
ers, they will, after an implementation period, lose some preference
rights and accept obligations to poorer countries. Alternatively, the
scheme could be designed to include a ‘one-way’ provision so that
free trade would monotonically increase in a dynamic world where
rankings change.
Concluding remarks
Developing countries have been understandably reluctant to com-
mit themselves to large reductions in their tariff levels. They are
concerned that open borders will lead to a flood of cheap imports
from more efficient producers, which could destroy their fledgling
industries before they have a chance to develop. Because they are
already characterized by high unemployment and weak private and
social insurance, many developing countries believe that the adjust-
ment costs from significant MFN tariff reduction are too large to be
seriously considered. Consequently developing countries have not
offered large reductions in border protection in WTO negotiations.
As a result, South–South trade has suffered, and developing coun-
tries have little bargaining power in their negotiations with devel-
oped countries.
The proposal in this chapter distributes new market access pro-
gressively, ensuring that the largest gains accrue to the smallest and
poorest countries, and it distributes liberalization obligations pro-
gressively, requiring that the largest and richest countries liberalize
most. There are many other issues associated with tariff reduction in
poor countries, including adjustment costs and declining revenue to
governments. These issues, and their implications for the need for
16
The Cotonou Partnership Agreement between the African, Caribbean, and Pacific (ACP) countries and
the European Union, signed in June 2000, mandates the negotiation by 2007 of a series of economic part-
nership agreements (EPAs) between regional groups of ACP states on the one hand and the EU on the
other.
technical and development assistance to poor countries, are not

discussed here. The advantage of MAP is that it provides significant
liberalization, does not demand reciprocity from poor nations to
richer ones, and places simple and well-defined obligations on both
rich and poor countries alike.
Special and differential treatment in rules
In the previous chapter we suggested that the WTO’s ‘policy space’ be

limited to areas in which there are clear international spillovers or
development gains to justify the prioritization of multilateral disci-
plines over national-policy autonomy. There is now a recognition that
many WTO agreements which impose ‘behind the border’ rules may
have gone too far in the direction of restricting the autonomy of devel-
oping countries to determine their own industrial policies. The
WTO’s Agreement on Subsidies prohibits export subsidies for all but
least developed countries. The Trade Related Investment Measures
(TRIMS) Agreement prohibits the use of a number of investment-
performance-related measures that have an effect on trade, including
local content and sourcing requirements. Both of these policies were
widely used by East Asian economies during their rapid industrializa-
tion in the last century. Further restrictions have been imposed by the
WTO’s TRIPS Agreement, negotiated in the 1986–94 Uruguay Round,
which introduced intellectual property rules into the multilateral
trading system. The TRIPS Agreement has considerable implications
for technological and industrial policy. By strengthening intellectual
property rights in developing countries, it is likely to increase the roy-
alty payments demanded by technology holders there, and also to cre-
ate or reinforce monopolistic positions in small markets. It also
restricts reverse engineering and other important methods of imitat-
ive innovation, thereby limiting the ability of firms in developing
countries to reduce their technological disadvantage.
To prevent further incursions on their policy autonomy, developing
countries should demand exemptions from restrictive multilateral
rules. This includes proposals by developed countries in the Doha
Round to develop multilateral disciplines for national competition

and investment policies. Proponents of an investment agreement
would like binding global rules that allow foreign investors the rights
to enter countries without any conditions and regulations which do
not apply to domestic firms, i.e. to be granted ‘national treatment’.
This type of international investment agreement is designed to maxi-
mize foreign investors’ rights while restricting the authority, rights,
and policy space of governments. Similarly, the proposals in the Doha
Round to introduce multilateral disciplines in competition policy
would limit the ability of governments and local firms in developing
countries to take actions which are to their advantage.
Certainly the WTO can serve as a device for commitment to good
policies, but such commitments should be part of voluntary pluri-
lateral agreements rather than compulsory elements of the WTO’s
‘single undertaking’. Compulsory commitments should not be made
in areas where the purported gains are controversial and where
implementation and opportunity costs are high. And where the gains
are certain and the costs are low for developing countries (such as
allegedly in efforts to liberalize some aspects of foreign direct invest-
ment regulations), there is no need for compulsion, because develop-
ing countries can be persuaded of the benefits of taking unilateral
action. For example, in FDI policy, developing countries have been
persuaded of the benefits of more open foreign investment regimes.
In 1999 alone, there were 140 changes to FDI regulations worldwide;
131 of these liberalized conditions were for FDI (UNCTAD 2000).
Developing countries have seemingly acted responsibly in their own
interests, without the need for multilateral compulsion.
There is a growing recognition that the WTO needs to move away
from agreements which enshrine compulsory rules under the ‘single
undertaking’. Hoekman (2004) suggests dividing the WTO’s disci-
plines into ‘core’ and ‘non-core’ agreements. Core disciplines would
be required to be unconditionally accepted by all countries.
Developing countries could be permitted to pass over non-core WTO
rules on development grounds.17
17
Hoekman (2004) suggests a consultative multilateral mechanism within the WTO to authorize poli-
cies which deviate from ‘non-core’ commitments. This authorization would be based on an assessment of
the effectiveness and impact of such policies.
Such differential treatment would facilitate the design of trade

agreements which are more likely to promote economic develop-
ment. A blanket proscription against government subsidies to tech-
nology (industrial policies) is likely to have an adverse effect on
developing countries and, indeed, it is likely in practice to be unfair:
the United States conducts its industrial policy largely through the
military, which supports a wide variety of technological develop-
ments that eventually have important civilian applications. And it
is hard to conceive of a trade agreement that would prohibit the
development of such technologies through defense programs. (Even
the EU has complained about America’s use of defense expenditures
as a hidden subsidy for its aerospace industry.)
Thus, each provision of a trade agreement should be assessed for
its impact on development, and designed to ensure that develop-
ment is enhanced, employing where necessary provisions for special
and differential treatment. Moreover, the totality of the trade agree-
ment should be assessed to ensure that a fair share of the benefits (the
net incidence) accrues to developing countries.
7
Priorities for a
Development Round
3
The natural question to be asked by each developing country as it
enters trade negotiations is ‘What agreement would make the most
difference for us?’ There is a corresponding question for the developed
countries: ‘What is it that we could give, which is of most benefit to
the developing countries, and which has the least cost (or perhaps even
benefit) to us at the same time?’. The developed countries’ natural
response may be to demand a quid pro quo. But such a demand would
be to look at the current negotiation outside of its historical context.
The developed countries have to date received the lion’s share of the
benefits from previous trade negotiations. Accordingly, they ought to
be willing to do more for the developing countries in this round.1
With little progress on the issues of concern to developing
countries—non-tariff barriers, intellectual property, migration,
unskilled intensive services, and agriculture—and new demands in
areas of dubious benefit to the developing countries, it has been hard to
see how the developing countries can benefit significantly.2 Actually,
1
The problem, of course, is that political globalization has not kept pace with economic globalization:
issues of international trade agreements are seldom looked at through which the same kind of lens through
which we look at domestic legislation. In national economic debates, we do not demand that the poor give
up an amount commensurate with what they get. Rather, we talk about social justice and equity.
2
Anderson (2001) compares the benefits of developed country liberalization on developing countries
with the benefits of developing country liberalization on developed countries. As a proportion of GDP in
each group the benefits to developing countries of the former exceed the benefits to developed countries
from the latter by a factor of 6. Given this imbalance in payoffs it is not surprising that developed countries
are in such a strong bargaining position.
in several areas, there is significant scope for gains to the developing

countries. In market access what matters is not just the average tariff
rate,3 but also the structure of tariffs. Escalating tariffs, where there are
higher tariffs on more-processed goods than on those less-processed,
inhibit the ability of developing countries to increase their manufac-
turing capacities, especially in areas which might represent a natural
comparative advantage, such as food processing or textiles.
For some goods, particularly skilled labor-intensive industrial
goods, the fact is that average developed country tariffs are already
much lower than those in developing countries.4 In these areas, the
large reductions in tariffs by developing countries would have put
large strains on these same countries. Were they at full employment,
a strong argument could be made that they would nonetheless bene-
fit, if they are given enough time and resources to adjust. But the
speed of adjustment that is likely to be demanded, and the absence of
adequate resources to facilitate the adjustment, mean that develop-
ing countries may be significantly worse off.
There is an important asymmetry of power in the negotiations:
what developing countries do in opening up their markets to devel-
oped countries has a much smaller impact on the developed coun-
tries than the converse–what the developed countries do in opening
up their markets to the developing world. In short, the developed
countries themselves gain from liberalizing their own markets,
because they are able to adjust, and the disturbances posed to them
by the developing countries are small. The developing countries are
in a far more disadvantageous position—they will need assistance in
making the required adjustments, and they should be given a longer
time within which to adjust.
Accordingly, in this book we propose that:
1. all WTO members commit themselves to providing free market

access in all goods to all developing countries poorer and smaller
3
Average tariff rates—when weighted by the amount of trade—may be particularly misleading, since high
tariffs lead to little trade, and accordingly such high tariffs may be given little weight in the computation of
the average, even though they have a very distortive effect.
4
Average tariff rates on industrial goods imported into the OECD countries fell from around 40% in
1950 to 1.5% in 1998 (Hertel 2000). In spite of these low average rates, there are tariff peaks, many of
which adversely hurt developing countries, so that, provided these are addressed, developing countries do
have something to gain.
PRIORITIES FOR A DEVELOPMENT ROUND 109
than themselves. Thus all developing countries could expect free

access to all markets with (a) a larger GDP and (b) a larger GDP per
capita;
2. developed countries commit themselves to the elimination of
agricultural subsidies; and
3. the promise of market opening not be undermined by technical
provisions like rules of origin.5
In short, reciprocity should not be the central feature of these nego-

tiations, as they have been in the past.
There is one other aspect of the context in which trade negotia-
tions are currently occurring. What distinguishes developed from
less developed countries is not only the extent and nature of market
imperfections, but also labor and physical resources. Developing
countries are intensive in unskilled labor; their greatest shortage is
probably in the ownership of physical capital. Developing countries
are disproportionately in the tropics6 and, currently, are more
engaged in the export of commodities, including natural resources.7
Thus, they differ from the developed countries in the products that
they export and import, which is why decisions about which goods
and services to liberalize, and which should be subject to restrictions
on subsidies, can make a great deal of difference for the general equi-
librium incidence.
Finally, we should note the dramatic transformation of the global
economy. In the nineteenth century, the (now) advanced industrial
economies transformed themselves from agriculture into manufac-
turing. Today only 14 per cent of employment and output in the
United States is in manufacturing, and the proportion in Europe is not
much higher.8 Now, they are transforming themselves from manufac-
turing economies into service and knowledge economies. Meanwhile,
the developing world itself is divided into several groups: subsistence
agriculture (much of Africa); export agriculture (Brazil and Argentina);
and those breaking out of agriculture and becoming increasingly
5
In the discussion in future chapters we address the concerns that preferential liberalization is inhibited
6
by strict or complicated rules. See Gallup, Sachs, and Mellinger (1998).
7
See FAO (2003).
8
Developed countries’ share of world trade in manufactures has fallen from 90% in 1970 to 72% in
2000 (World Bank 2002a).
centered on manufacturing. For the agricultural exporters, of course,

the failure to liberalize trade in agriculture and to remove subsidies
has been particularly costly.9
There is, as a result, a fundamental tension in current trade nego-
tiations. The developed countries want to protect their declining
industries and to gain market access for their expanding industries.
But their declining industries are declining largely because of com-
petitive pressures from the developing countries. Hence, the sectors
that they are most interested in protecting are precisely the sectors
that are of the greatest concern to the developing world. It is not as if
America is mostly concerned with protecting itself against Europe,
or vice versa (though there is some element of that). And the sectors
that are declining are, by the same token, those with the lowest-
wage workers. Hence protection elicits concerns about equity and
social justice within the developed countries—but the failure to
extend these concerns to developing countries shows a particularly
narrow vision which is out of step with economic globalization.
At the same time, by demanding market access for the sectors
which are growing, the developed countries hope to catapult the
advantage that they already have—the first-mover advantage—into
a longer-term advantage. For that very reason, were such a strategy
accepted, it would inhibit the development transformation of the
poorer countries, making it all the more difficult for them to move
from traditional products to become effective competitors with the
more developed countries.
The chapters which follow present pro-development priorities
that should form the core of the Doha Round agreements. Much of
the recent discussion has focused on agriculture, but there is much
more to a true Development Round. Primary attention should be
given to market access for goods produced by developing countries.
There is an urgent need to reduce protection on labor-intensive
9
But note that most of the progress in trade negotiations during the past half century has focused around
liberalization of manufacturing (other than textiles)—the goods that are of diminishing importance to the
advanced industrial countries but of increasing importance to middle-income developing countries. There
is a certain irony: while the United States and Europe may have thought that they were negotiating trade
agreements that were of most benefit to themselves, in fact they negotiated a global trading regime that, if
it is fairly implemented (and setting the non-tariff barriers aside), is likely in the future to be of most benefit
to China and other middle-income developing countries.
PRIORITIES FOR A DEVELOPMENT ROUND 111
manufactures (textiles and food processing), and on unskilled services

(maritime and construction services). Priority should also be given to
the development of schemes to increase labor mobility—particularly
the facilitation of temporary migration for unskilled workers. As tariff
barriers have come down, developed countries have increasingly
resorted to non-tariff barriers as one remaining protectionist instru-
ment. These need to be circumscribed. The proposals are motivated
by empirical analysis of the gains and costs of liberalization. For ease
of exposition the analysis of this evidence is presented separately in
Appendices 1 and 2 and a summary of some of our conclusions is
briefly set forth in Table 7.1.
Table 7.1. Development issues in the Doha Round
Issues Context for developing Countries Doha: commitments made and Agenda for a true Development Round
subsequent progress
Labor migration and More important than capital flows, especially Largely ignored at Doha and little progress Development of provisions for temporary
unskilled labor-intensive for developing countries. since. WTO members are currently submitting movement of unskilled workers and the
services Liberalization of services has hitherto focused proposals regarding both the structure and the facilitation of remittance flows.
on skilled intensive services. contents of the new negotiations—the vast Priority to the liberalization of unskilled labor-
majority of proposals concern the intensive services.
Migration discussion has focused on the liberalization of skilled labor-intensive services.
movement of high-skill rather than low-skill
workers.
Agriculture Huge developed-country subsidies and Clear expectation at Doha that protection Eliminate production as well as export
protection. would be reduced. subsidies, focusing first on commodities
Since Uruguay Round, subsidies increased Since Doha, EU and US have offered to reduce whose benefits to producers far exceed costs
rather than reduced. export subsidies, but little progress in to consumers.
Uruguay Round created a distinction between eliminating production subsidies. Provide assistance to consuming nations.
trade-distorting and non-trade-distorting Preliminary ruling against US cotton subsidies In interim allow countervailing duties.
subsidies; allegedly non-trade-distorting weakens US bargaining position.
subsidies still hurt developing countries.
Industrial goods Existing tariff structures (including tariff peaks) Doha committed nations to reducing tariffs ‘in Eliminate all tariffs against least developed
discriminate against the goods of interest to particular on products of export interest to countries.
developing countries. developing countries’. Eliminate tariff peaks and tariff escalation.
Tariff escalation (higher tariffs on processed Recommitment to same principles in 2004, Reduce tariffs on industries where developing
and semi-processed goods) restricts industrial but little concrete progress on modalities. countries have natural comparative advantage:
diversification in developing countries. labor-intensive goods and agricultural processing.
Non-tariff barriers As tariffs have been reduced, developed Doha referred to competition issues of interest Eliminate dumping duties—establish a single
countries have put increasing reliance on non- to developed countries, but made little fair competition regime for domestic and
tariff protectionist measures; implementation reference to concerns of developing countries. foreign producers.
discriminates especially against developing US has used safeguard measures on steel, but Ensure that developing countries can
countries. rejected by adverse WTO ruling. subsidize infant industries and offset high
Increasing use by both developed and less interest rates without countervailing duties.
developed countries represents threat to
liberalized trade regime.
Development box Developing countries’ attempts to use The Doha Declaration recognized the need for Developing countries to use measures that are
domestic policies to promote development ‘Special and Differential’ treatment (STD) of in their development interests even if
should not be circumscribed by the WTO. developing countries. proscribed for developed countries—
However, STD as currently envisaged amounts particularly actions to protect poor farmers.
to little more than longer implementation Ensure food security.
periods and some broader exemptions for the
poorest nations.
Intellectual property Access to advanced technology important for Doha promised a more balanced intellectual Pro-generic-drug policy.
development. property regime, with special attention to Compulsory licensing for any life-saving
health issues. medicine.
Lack of access to life-saving drugs,
jeopardizing lives. Some progress on compulsory licensing, but Higher novelty standards and more restrictions
problems remain. on scope, especially in relationship to
Biopiracy—western companies patenting
traditional foods and medicines. Little progress on biodiversity and other issues. threatened bio-piracy.
Bilateral agreements restrict access to generic
drugs, showing lack of concern for welfare of
Restrictions on tax Developing countries compete with tax Not addressed. Increased transparency of subsidies to foreign
concessions concessions to attract foreign businesses; net firms.
beneficiary is international businesses. Restriction of tax concessions.
Arms sales and corruption Have major deleterious effect on Not addressed. Prohibition of arms sales.
development. Criminalization of bribery; allowing tax
deduction only for royalty payments that are
‘published’.
Table 7.1. continued
Issues Context for developing Countries Doha: commitments made and Agenda for a true Development Round
subsequent progress
Elimination of secret bank accounts.

Commitment to repatriate corrupt funds.
Dispute resolution and fairer Dispute system favors rich countries and is Not on agenda. Multilateral enforcement.
mechanisms for enforcement under-utilized by developing countries. Monetization of sanctions (auctioning off right
Asymmetries in enforcement: trade sanctions to sanction).
by small developing country unlikely to have Expansion of technical assistance to ensure
major effect. that developing countries have access to equal
protection under the WTO’s dispute
settlement system.
Extension of unilateral Cost to developed countries of opening Doha recognized principle of special and Extend to more countries and eliminate
disarmament markets would be small; benefits to differential treatment recognized. restrictions which limit its impact.
developing countries would be large. Europe has already adopted Everything but
Arms initiative, although rules of origin limit
benefit.
US has adopted AGOA for poorest African
countries.
Institutional reforms Bargaining process lacks fairness, Doha hinted at reforms to processes. Replacement of ‘green room’ procedures by
disadvantages the poor, and is undemocratic. Since then there has been slight progress in ‘principle of representativeness’.
Impacts on developing countries are never transparency. More openness and transparency.
assessed before adoption. Creation of evaluation/research unit to assess
impacts of proposals, as well as bilateral and
regional agreements to determine whether
they ‘create’ or ‘divert’ trade.
8
How to Open up Markets
3
The general argument in favor of trade liberalization is that it allows
the expansion of the size of markets, allowing the global economy to
take further advantage of the economies of scale (the argument Adam
Smith put forward more than two-hundred years ago), and it
enhances global efficiency in production and exchange. Thus trade
liberalization must be managed carefully to ensure that developing
countries benefit from it and are not left worse off. In this chapter, we
discuss the most important market access issues in the Doha Round
and we show what must be done to promote development. As noted
earlier, the standard argument that trade liberalization necessarily
makes all countries better off (though not necessarily all individuals
within each country) is predicated on a set of assumptions that is not
satisfied in most developing countries: full employment, perfect
competition, and perfect capital and risk markets.
Labor mobility and unskilled

labor-intensive services
The General Agreement on Trade in Services (GATS) recognizes four

modes of service delivery. The temporary movement of natural
persons (so called Mode 4 service delivery) has received by far the
smallest attention in terms of the volume of scheduled concessions.

Yet differences in factor payments across countries provide evidence
that factor movements would substantially increase global product-
ivity. If factor payments equal marginal products,1 then the largest
discrepancies are associated with the payments to unskilled labor,
then to skilled labor, and lastly to capital. Accordingly, agreements
that provide for the mobility of unskilled labor would do most to
increase global efficiency.
Yet despite the tremendous development potential of this
reform, the limited progress that has been made in this area has
been largely associated with the intra-corporate movement of
skilled personnel—an issue of interest to developed countries.
Thus far Mode 4 has not progressed in a way that allows develop-
ing countries to use their comparative advantage in low- and
medium-skill labor-intensive services. Nor has enough attention
been given to proposals to facilitate remittances—the payments
from migrant workers back to their families in developing coun-
tries. Governments have a role to play in maximizing both the
value of remittances and their impact on development. Efforts to
formalize the structure of remittance flow (much of which cur-
rently moves through informal channels) could make it easier,
safer, and cheaper to transfer funds. For example, governments
could ensure that migrants have access to secure and low-cost
financial services and could regulate remittance-handling inter-
mediaries to prevent malpractices. As well as increasing the flow
of remittances, remittance policies can improve the development
impact of remittances at the receiving end. For example, micro-
finance and micro-enterprise support initiatives have encouraged
remittance-receiving clients (especially small businesses) to
access credit and savings accounts.2 Finally, the further develop-
ment of remittance-backed bonds could help liquidity-constrained
1
They may not, and the disparity between factor payments and the value of marginal products may dif-
fer across countries, if the degree of market imperfections differs.
2
For an example of an initiative in this area see the case of the financial institution PRODEM in Bolivia,
which focuses on the promotion of savings and the offer of new financial services to remittance receivers.
See UNDP (2003b). A number of best-practice scenarios from Latin America and Asia were presented and
documented in the November 2000 ILO conference in Geneva on ‘Making the Best of Globalization:
Migrant Worker Remittances and Micro-Finance’.
HOW TO OPEN UP MARKETS 117
developing countries to use future flows of remittances to raise

external finance relatively cheaply.3
As well as facilitating the movement of natural persons (Mode 4),
there is scope for liberalization of other service industries of import-
ance to developing countries. Services account for, on average,
50 per cent of developing countries’ GDP, but developing countries
account for only 25 per cent of the world’s services exports. While the
last decade has seen considerable liberalization of high-skill services,
there has been less progress in those unskilled-labor-intensive
services of interest to developing countries.
The liberalization of services could yield significant welfare gains
to developing countries—indeed a large number of empirical studies
suggest that service sector liberalization has the potential to deliver
larger gains than agricultural or manufactured goods.4 The estimates
are large because protection levels are high in the service sector, and
services make up a large (and growing) share of world trade.
Additionally, services are key inputs into the production of almost
all goods.
However, the large predicted gains from service sector liberaliza-
tion have to be weighed against the relative complexity of service
sector reform, where the identification and elimination of trade bar-
riers is significantly more difficult than in merchandise trade. In par-
ticular, three concerns commonly arise. First, there is a view that
only a small fraction of service sector reform is actionable through
WTO negotiations. Second, several important elements of the
reform agenda (particularly liberalization of restrictions on foreign
direct investment within Mode 3 service delivery5) are successfully
progressing through unilateral policy changes outside the WTO.
Third, multilateral commitments within the WTO are seen by
several developing countries as a particularly blunt instrument of
3
In 2001, Banco do Brasil issued US$300m worth of bonds through Merrill Lynch using the future yen
remittances from Brazilian workers in Japan as collateral. The terms of these bonds were more favorable
than those available on sovereign issues (with a BBB⫹ Standard and Poors rating compared to BB⫺ on
Brazil’s sovereign foreign currency rating). For a review of securitization of remittance flows see Ketkar and
Ratha (2000).
4
See, for example, Hertel et al. (2000), Dee and Hanslow (2000), Brown, Deardorff, and Stern (2001),
Francois, van Meijl, and van Tongeren (2003), Verikios and Zhang (2004).
5
Mode 3 refers to trade in services delivered through foreign commercial presence, particularly foreign
subsidiaries of multinational firms.
reform, lacking the flexibility to deal with the country-specific

implementation challenges thrown up by liberalization in the ser-
vice sector. Together these concerns contribute to the continued low
priority given to GATS commitments within the overall agenda.
This is unfortunate because it undervalues the significant and grow-
ing service export interests of developing countries and also because
it has drawn attention away from those policy proposals (scattered
throughout the agenda) which could facilitate and improve the effect-
iveness of unilateral reform in developing countries.
Many developing countries have large and growing export inter-
ests which could be pursued in the Doha Round. The substantial
growth in offshore outsourcing (Modes 1 and 2)6 has led to high
growth rates of exports from developing countries in particular busi-
ness services7 and ICT, but also in health, education, and audiovisual
services. Barriers to trade in these areas include national authoriza-
tion, local authentication requirements, and regulatory standards.
There is significant scope for liberalization in Mode 1, which lags
behind Mode 3 in terms of both the number and scope of commit-
ments. The Uruguay Round left many areas of Mode 1 trade without
bound commitments. A large proportion of commitments provided
only partial market access (60 per cent in legal services; 78 per cent
in voice telephone services; 41 per cent in accounting; see Matoo and
Wunsch 2004). In several areas where developing countries have a
comparative advantage there is a case for broad formulaic rules in
favor of national treatment and increased market access.
At the same time there are other non-market access reforms
which could complement service sector reform and increase the
benefits available to developing countries. The tourism sector (Mode
2) is one of the most important sources of foreign exchange for many
developing countries. While the sector is generally quite liberal in
terms of government restrictions,8 developing countries suffer from
rampant anti-competitive activities with the industry (centered in
6
Service delivery through Modes 1 and 2 refers to cross-border supply and consumption abroad
respectively.
7
In India exports of business services grew by 43% between 1995 and 2000 (Matoo and Wunsch
2004).
8
There has been a high number of commitments in major tourism sectors, in particular hotels and
restaurants (123 members). See WTO (1999).
the North) which minimize spillover and multiplier effects. In this

and other areas (for example maritime transport), an effective multi-
lateral anti-trust framework could deliver large gains to developing
countries and support further unilateral liberalization.
The same is true in Mode 3 liberalization, where the enthusiasm
for FDI in the cross-country empirical literature is tempered by negat-
ive experiences at the national level. While it is true that the unilat-
eral liberalization of restrictions on foreign investment continues
apace without multilateral action, there are nonetheless several
opportunities for WTO action which could increase the benefits that
developing countries derive from the liberalization of FDI. For example,
developing countries’ experiences with FDI could be improved by
agreements to limit the adverse consequences of competition for
investment through fiscal and financial incentives9 and also by agree-
ments to facilitate anti-trust enforcement actions.
Another problem is that in many cases, the ramifications of
service sector reform extend beyond the impacts on trade. Inevitably,
then, debates about service sector liberalization devolve into funda-
mental debates about national economic and social policy. Is it right
that the media, for instance, be controlled by a few rich foreign
firms, who are able to use their wealth to control the flow of infor-
mation to the citizenry? A further concern is that many service sec-
tor liberalizations might have social consequences for the poor, for
example by increasing prices of essential services or by reducing
access. Opening up markets has been accompanied at times by a
reduction in competition, and an increase in prices;10 in the case of
financial services, there are even allegations that the supply of credit
to medium and small domestic enterprises has been reduced. Private
firms may be less willing to engage in cross-subsidization of market
segments in poor and rural areas. Even if liberalization leads to lower
average costs through increased competitiveness and efficiency,
prices for some end-users may rise. The WTO could promote service
9
For a discussion of harmful tax practices see OECD (1998). For welfare losses from international tax
competition see Charlton (2003). See also discussion in chapter 9.
10
For example, privatization of utilities—such as South Africa’s experience of granting its newly privatized
telecommunications utility Telekom a 5-year monopoly—can lead to inefficient services. Similarly the poor
regulation of financial sectors across South-East Asia contributed to instability prior to the crises of the late
1990s. Poor electricity deregulation has led to problems in many countries.
sector liberalization by acting to mitigate (or at least not exacerbating)

these concerns through other parts of the agenda. For example, regu-
latory agreements which constrain the ability of governments to
avail themselves of appropriate industrial, social, and redistributive
policies might reduce the incentive for governments to engage in
liberalization programs which entail adjustment costs.
Service sector reform thus offers large benefits to developing
countries but is not receiving commensurate attention in the Doha
Round. There is much to be done within the WTO to unlock welfare
gains from service sector reform, including pursuing the developing
countries’ market access agenda in labor-intensive and outsourced
services and promoting reform in other parts of the agenda to
amplify the benefits of service sector liberalization and limit its
costs.
Agriculture
Chapter 3 highlighted the persistently high levels of agricultural

protection in the OECD.11 Yet agriculture is crucial to developing
countries. It represents almost 40 per cent of their GDP, 35 per cent
of exports, and 70 per cent of employment.
Because agriculture is such an important part of both national eco-
nomic development and daily livelihoods in developing countries,
agricultural reform must proceed carefully. Agricultural liberaliza-
tion presents developing countries with the benefits of increased
market access, but also the (potential) costs of higher prices for
domestic consumers. The fundamental point is that consumers bene-
fit from lower prices that result from large agricultural subsidies, and
producers lose.12 The producers are typically poor farmers, often far
worse off than the urban net consumers. Given the limited capacity of
11
Total OECD spending on agricultural subsidies is more than US$300bn per year. This is almost six
times the total aid from OECD countries to all developing countries (US$50–60bn per year).
12
There is another reason to be wary of an excessive focus on agriculture. Development requires less
developing countries to move into sectors with higher rates of potential productivity improvements, to
develop their dynamic comparative advantage, not just their static comparative advantage.
developing countries to effect redistributions, there can be a signifi-

cant welfare loss from such adverse distributional impacts. The net
effect of wide-ranging agricultural reform varies across developing
countries depending on the composition of their exports and imports
of different commodities, and the price sensitivity of those com-
modities to liberalization. The potential for losses highlights the
need for a more fine-grained approach which would differentiate
among crops and countries, and emphasizes the importance of
adjustment assistance, which would need to vary among developing
countries, depending on the magnitude of the adverse impact.13
The WTO should focus on liberalizing those commodities which
have the largest positive effect on producers and the smallest adverse
consumption effects. One determinant of the net effect of this kind
of reform is the level of protection for each commodity and the con-
sequent impact of liberalization on prices.14 Another important
determinant of the welfare effects of liberalization is the agricultural
trade balance across countries. There is a division between temper-
ate products (some types of crops and livestock), where developing
countries are largely net importers and developed countries are
largely net exporters, and tropical products, for which developing
countries are largely net exporters. Most developing countries are
net importers of program crops,15 which are precisely the commodit-
ies that have the highest domestic support and stand to experience
the largest price increases. It is therefore not surprising that most
studies predict that most developing countries are worse off as a
13
In addition, countries which are importers of subsidized commodities as well as producers should be
allowed to impose countervailing duties. Such duties would simultaneously enable producers to receive
prices that would correspond more closely to what they would have received in the absence of the distor-
tionary subsidies in the advanced industrial countries and provide the revenues with which these countries
could protect consumers from the adverse consequences of the price increase. Moreover, since those in
the advanced industrial countries would receive less benefit from their distortionary subsidies, such a
reform might reduce political pressures for the subsidies.
14
There are large differences in the extent to which different agricultural crops are subsidized. Tariffs are
particularly high in the feed grains, dairy, and food grains sectors, while dairy products, meat, and livestock
are the world’s most subsidized exports. Producer payments are highest for grains and oilseed sectors and
lowest for meat, livestock, and dairy (Hertel et al. 2000).
15
This includes Mexico, ‘Rest of South America’ (a regional average which excludes Argentina and
Brazil), China, Indonesia, South Korea, ‘Rest of South Asia’ (a regional average which excludes India),
Tanzania, Zambia, ‘Rest of Sub-Saharan Africa’ (a regional average which excludes Tanzania and Zambia),
and the average of the Middle East and North African Countries. Brazil, India, Argentina, and Vietnam are net
exporters (Dimaranan, Hertel, and Keeney 2003).
result of the terms-of-trade effects following this kind of reform.

Indeed Dimaranan, Hertel, and Keeney (2003) find that gains accrue
primarily to developed countries in the Cairns Group as well as the
two largest developing country exporters, Argentina and Brazil.
These countries are the strongest advocates for the existing agricul-
tural reform agenda. Still, it is possible that, as producer prices
increase, some developing countries will switch from being net
importers to net exporters.
The existence of net losses for developing countries in some areas
of reform should not imply that no reform is required—rather it
suggests that a selective and gradual approach is needed and that
considerable adjustment assistance may be required. The most
important subsidies to eliminate would be those where the con-
sumption benefits of the current subsidies are small relative to the
cost to producers. Attention should be focused on the elimination of
tariffs and quotas on tropical products, processed foods, and other
commodities which developed countries export or for which they
have high export elasticities with respect to price. Elimination of
cotton subsidies would raise producer prices for cotton, but have a
small effect on standards of living in developing countries as a result
of the small increase in the price of cloth. Similarly, subsidies for
crops which are disproportionately consumed by the wealthy will
have the least adverse distributional effects.
Furthermore, the potential adverse effects of agricultural liberal-
ization on large segments of society suggest the importance of a
gradual approach, allowing urban workers time to adjust. It would
also be desirable for developed countries to give some of the money
they previously expended on subsidies to assist the developing
countries in the transition.
Appendix 1 reviews the empirical evidence on the effect of agri-
cultural liberalization in OECD countries on various regions of the
world. Clearly more research needs to be done to identify the precise
effects of liberalization on individual poor countries. The studies
surveyed in Appendix 1 indicate that uniform elimination of all agri-
cultural protection could result in negative terms-of-trade shocks for
some of the poorest developing countries and sharp declines in farm
incomes in Europe and North America. The latter are in a position to
bear the costs, especially given the large savings from the elimination
of subsidies: the former may not be. A reform agenda must carefully
discriminate between liberalization instruments and targets. Such an
agenda would have three key components.
First, there should be a significant reduction in border protection
in developed countries (particularly the EU), including tariff cuts and
the elimination of export subsidies. Tariffs on the goods produced
primarily by developing countries as well as those consumed pri-
marily in developed countries should be reduced most rapidly. For
example, the elimination of US and EU quotas and tariffs on sugar
and tropical products would increase the price received by develop-
ing world producers but only have a small effect on consumer prices
in developing countries. Similarly, the elimination of cotton subsi-
dies would have only a small effect on consumer prices in developing
countries.
Second, domestic production support for price-sensitive neces-
sities that are widely consumed in developing countries should be
reduced gradually, with some of the savings in developed country
subsidy budgets being directed at ameliorating the adjustment costs
of those in the developing world. Many developing countries in
North Africa, Sub-Saharan Africa and Latin America (though not
Brazil, Argentina, or Mexico) rely on imports of subsidized grains
and oilseeds from OECD producers. The empirical evidence
reviewed in Appendix 1 suggests that these countries are particu-
larly exposed to agricultural reforms which might increase the price
of some commodities.
Third, domestic support should be shifted from market price sup-
port to alternative payment systems. Reinstrumentation of protec-
tion in OECD countries towards the least trade-distorting
instruments (such as land-based payments) is one possible means of
compensating OECD farmers while minimizing the impact on
developing world consumers. But many of the so-called non-trade-
distorting subsidies do in fact lead to increased production, and too
much has been made of the distinction between export subsidies and
production subsidies. The WTO makes a distinction between
explicit export subsidies and other forms of domestic subsidies, yet
both types of payment can increase production and exports and
depress world prices.16 Since domestic subsidies are treated more

permissively in the WTO, several OECD countries have reduced
their export subsidies and increased their direct domestic support
payments to comply with their WTO commitments. In the US and
EU, the annual values of export subsidies for cereals and beef
declined by US$4.1 billion between 1990 and 1998–9. In the same
period, domestic support in the form of exempt direct payments for
those commodities rose by an estimated US$18.9 billion a year in
the European Union alone (ABARE 2001). However, the trade effects
of various types of domestic subsidies are often understated. While
the impact of export support on developing countries per dollar of
subsidy is greater than production-based support, the difference is
small if the elasticity of demand is small, which is the case for many
agricultural commodities. Even non-production-based support
(‘decoupled’ payments primarily in the ‘Green Box’) have an impact
on output and prices. These payments favor OECD producers by
providing them with cheap (or free) credit to use potentially for
investment and expansion of production. The distinction between
trade-distorting subsidies and non-trade-distorting subsidies is
based on a particular economic model, in which capital markets are
perfect. Trade-distorting subsidies are then subsidies which change
the marginal return to production or which reduce the marginal cost
of production. Thus generalized income supports in this view are not
production-distorting, nor are payments to keep land fallow. But
both of these may, in fact, be production-distorting if, for instance,
farmers face credit constraints. Then, in effect, the subsidies provide
additional finance which allows farmers to expand production.
Liberalization of industrial goods
While average developed country tariff rates are low, developed

countries maintain high barriers to many of the goods exported most
16
The WTO classifies domestic subsidies according to their distortionary effect on trade: amber (directly
trade-distorting); blue (indirectly trade-distorting production payments); green (non-trade-distorting).
intensively by developing countries (US$31 billion). When weighted

by import volumes, developing countries face average manufactur-
ing tariffs of 3.4 per cent on their exports to developed countries,
more than four times as high as the average rate faced by goods from
developed countries, 0.8 per cent (Hertel and Martin 2000).17
Moreover, aggregate data hide the existence of tariff peaks (discussed
in Chapter 3). OECD tariffs are particularly high for goods of impor-
tance to poor countries, such as low-skill manufactures (especially tex-
tiles) and processed foods. For example, in 2001, clothes and shoes
accounted for only 6.5 per cent of US imports in value terms but they
brought in nearly half of the US$20 billion of US tariff revenue. More
tariff revenue was collected by the US government on the import of
shoes than on the import of automobiles, even though the value of shoe
imports is around one tenth of the value of automobiles.
Comparisons of nominal tariffs do not fully represent the distortion
caused by escalating tariff structures, and do not provide information
about the impact of tariffs on the value-added of processed products.
Thus nominal tariff levels tell us little about the real trade impact of
the tariff escalation, and to look more closely at the trade effect we need
to examine the ‘effective tariff rate’. The effective tariff rate is a func-
tion of the tariffs assessed on the component parts and the final prod-
uct, the technological process involved, and the relative prices of all
inputs into the final product. Analysis of tariff escalation using effect-
ive tariff rates demonstrates the effect of the existing tariff structures in
many developed countries on the industrial development of poorer
countries. By imposing higher tariffs on the output of manufactured
goods than on primary products, developed countries are in effect
imposing significantly higher trade taxes on manufacturing value
added in developing countries. Such tariff ‘escalation’ serves to dis-
courage the development of, for example, food processing in less devel-
oped countries since the effective tariff rate on value added in food
processing is very high. Such tariff peaks and tariff escalation are mani-
festly unfair and have a particularly pernicious effect on development
by restricting industrial diversification in the poorest countries.
17
The distortion is even larger if one recognizes that the quantities imported are reduced as a result of
the high tariff barriers. (In the measure cited, a prohibitive tariff would have no weight in the measure, since
there would be no imports.)
A second reason why developing countries should be pushing to

have industrial tariffs given high priority in the Doha Agenda is that
barriers to South–South trade are quite high. The average import-
weighted tariff on the exports of manufactured goods from develop-
ing countries to developing countries is 12.8 per cent (Hertel and
Martin 2000). Anderson et al. (2000) estimate that the welfare gains
to developing countries derived from the liberalization of trade in
manufactures by other developing countries is US$31 billion.
Non-tariff barriers
It is not surprising that as tariffs have come down, non-tariff barriers

have assumed increasing importance. Trade agreements may do lit-
tle to alter protectionist sentiment—and the politics of special inter-
ests. They do, however, change the form that such protection can
take. Just as developed countries have discriminated against devel-
oping countries in the structure of their tariffs, so too many of their
non-tariff barriers have particularly adverse effects on developing
countries.
Developing countries have repeatedly found that as they make
inroads into a market in the United States or Europe, they are
slapped with dumping duties or face some other form of non-tariff
barrier. Though ostensibly the Uruguay Round marked the end of
the so-called voluntary export restraint, the United States has talked
about reinstating such restraints against China. The effect of these
non-tariff barriers is far greater than indicated by the actual duties
imposed. The fear that they will be imposed has a chilling effect on
development: it increases the risk associated with investing in an
export-oriented industry, which is particularly important in
economies already facing high interest rates. Often initially high
duties are imposed only to be revised down substantially but the ini-
tially high duties suffice to drive the exporting firm out of business.
Some solution to the problems posed by non-tariff barriers should be
high on the agenda of any development round.
There are four important categories of non-tariff barriers: (1) dumping

duties, which are imposed when a country (allegedly) sells products
below cost; (2) countervailing duties, which can be imposed when a
country subsidizes a commodity; (3) safeguards, which can be imposed
temporarily when a county faces a surge of imports; (4) and restric-
tions to maintain food safety or avoid, say, an infestation of fruit
flies. The advanced industrial countries have used all of these at
times to restrict imports from developing countries when the latter
have achieved a degree of competitiveness which allows them to enter
the markets of the developed countries. Many of these measures
are described as ensuring ‘fair trade’, but from the perspective of
developing countries, they ensure ‘unfair trade’. They are evidence
of the hypocrisy of the North. Increasingly, however, developing
countries are using such measures against each other and against the
advanced industrial countries, and in that sense they represent a
hidden threat to a trade liberalization regime.
There has been a large increase in the number of anti-dumping
claims. Between 1995 and 2002, 2,063 dumping cases were initiated.
The US (279) and the EU (255) were two of the largest initiators. It
does not seem sensible that the countries with the largest capacity to
absorb shocks and compensate import-competing interests should
be the most common users of anti-dumping laws.
The problem is that dumping has been used as a safeguard measure
when there is a separate safeguard measure for this purpose. There is
a reason for this: safeguard measures provide for only temporary pro-
tection, to assist in adjustment, while dumping can provide longer-
term protection.
Part of the problem with the schemes is how they have been
implemented. Consider, for example, America’s use of dumping
duties. The accused must respond in a short period of time to a long
demand for information (in English), and when the accused is unable
to do so, the US government acts on the ‘best information available’
(BIA), usually the information which has been provided by the
American company trying to keep out its rivals. High initial duties
are imposed, which regularly get revised downward when better
information becomes available. But meanwhile, long-term damage
has been done, as American buyers will not purchase the commodity
given the uncertainty about the level of tariffs they may have to
pay.18 America’s provisions for dumping duties (and in some cases
countervailing duties) for China and some of the former communist
countries have been particularly egregious. In the ‘surrogate country
methodology’ which is used to assess the cost of production (the
benchmark against which charges of dumping are assessed) costs of
production are compared with those of a ‘similar’ country. In one
instance, the United States used Canada as the country most similar
to Poland. Not surprisingly it was found that the costs of production
were high, justifying a high dumping duty.
Safeguards are another form of non-tariff barrier. A WTO mem-
ber may take a ‘safeguard’ action (i.e. temporarily restrict imports
of a product) to protect a specific domestic industry from an
increase in imports of any product which is threatening to cause
serious injury to the industry. Recent years have seen a dramatic
increase in the use of safeguards around the world: the incidence of
safeguards has risen from 2 in 1995 to 132 in 2002. The increase is
a cause for concern because many of the safeguard measures imple-
mented by WTO members are not consistent with WTO rules.
Indeed, all safeguard measures examined by WTO panels and the
Appellate Body have so far been found to be inconsistent with those
rules.
Safeguard measures have been available under the GATT (Article
XIX). However, they were infrequently used. The Agreement on
Safeguards clarifies and reinforces Article XIX. It sets forth criteria
for the application of safeguard measures: (1) the product is being
imported in increased quantities, absolutely or relative to domestic
production; (2) the product causes or threatens to cause serious
injury to domestic industry; and (3) the safeguard measure shall only
be applied to the extent necessary to prevent or remedy serious
injury and to facilitate adjustment.
The safeguard measure has probably been underused by develop-
ing countries and has certainly been overused by the United States.
American safeguard legislation, for instance, makes insufficient dis-
tinction between industries which are declining because of trade
18
For a more complete description of these abuses, see, for instance, Stiglitz (1997).
and those which would be in decline even in the absence of trade

liberalization.
The US Presidential Proclamation of 5 March 2002 imposed safe-
guard measures on ten steel product groupings in the form of addi-
tional tariffs up to 30 per cent. Almost immediately, several
countries, including the EC, Japan, South Korea, China, Switzerland,
Norway, New Zealand, and Brazil, engaged WTO dispute settlement
procedures against these measures. These countries argued that
none of the US measures had been taken as a result of unforeseen
developments, as required under the WTO rules. For most products,
imports had not increased; for all products but one the US had not
properly established the necessary causal link to the alleged serious
injury suffered by the US steel industry; the US had exempted
imports from Canada, Mexico, Israel, and Jordan from the measures
in a manner inconsistent with the WTO rules. The WTO’s panel sub-
stantially agreed and concluded that each of the US measures was in
violation of the WTO rules. President Bush’s action in imposing
steel tariffs exemplifies the misuse of safeguards.
While the argument for safeguard measures is persuasive, they have
been widely abused, especially by the developed countries. If the rich-
est country in the world, the United States, with a strong safety net,
relatively high employment level, etc. has to resort to safeguard mea-
sures to protect itself against a surge of imports, how much more jus-
tified are developing countries in imposing such measures. Indeed, it
is hard to conceive of many important liberalization measures against
which safeguard protections could not justifiably be invoked by devel-
oping countries. This highlights again the need to set clearer standards
at the international level. For instance, for a safeguard measure to be
imposed, the country should have to show that there is injury but that
it is substantial, entailing a loss, say, of at least 1 per cent of the jobs in
the country, and that the burden on the country’s social safety net is
such that it would be hard pressed to absorb it. The threshold standard
should be lower in developing countries. Such a reform would ensure
that the safeguard measures would only be used in cases where trade
disturbances imposed significant adjustment burdens.
There are three reforms that would make a great deal of difference.
The first is to recognize the principle of national treatment: in
addressing problems of unfair trade, the legal framework should be

the same for domestic firms as it is for foreign firms. In the case of
dumping, for instance, firms are charged with selling below cost. To
an economist, the natural question is, ‘Why would a firm ever sell
below marginal cost (and what is the relevant economic concept)?’
The answer is to try to drive out rivals, to establish a monopoly or
dominant position in a market that would enable it later on to sell at
a high price, well above costs. Thus American anti-trust law, in
assessing whether predatory pricing (the equivalent of dumping in a
domestic context) has occurred, attempts to assess whether price is
below the relevant cost, and whether there is evidence that it is
likely that the accused will recoup his loses. As a result of this high
standard, few cases of predatory pricing have been successfully pros-
ecuted. Subjecting foreign firms to the same standard would ensure
that dumping charges were being used to preserve competition, not
to reduce the threat of foreign competition. (The double standard is
highlighted by the fact that if American firms were subject to the
standard used in a dumping case, a large proportion of American
firms would be found guilty of dumping.)
The second is to create a new international tribunal as the first
‘court’. Today when, for instance, the United States accuses firms of
a foreign country of dumping, it acts as prosecutor, judge, and jury.
Though the process is governed by a ‘rule of law’, in the sense that
there are well-defined procedures, the process often operates in a
highly unfair way. There is a costly and lengthy WTO process which
can be, and has been, used to rectify gross abuses, as in the case of the
US-imposed steel tariffs. But it would be far better if the original
decision was taken out of the hands of the country and put into those
of a specialized international tribunal.
The third is that the implementation legislation and practices of
countries should be reviewed to ascertain whether they are is fair
and non-discriminatory, both de jure and de facto, and are in confor-
mity with widely accepted economic principles. An example already
referred to is the use of BIA. As another example, almost all economists
agree that the relevant cost concept for judging dumping is marginal
not average costs, yet legislation in many countries uses average costs.
This means that dumping charges are often sustained in cyclical
industries, in downturns, where marginal costs are considerably

below average costs.19
The determination of whether subsidies have been provided is
another example which has been highly contentious. A Development
Round should clarify this, in ways which ensure that governments
may undertake industrial policies to promote nascent industries. This
is particularly important because the form of subsidy in countries like
the United States—e.g. research in the defense industry, the benefits of
which spill over to civilian uses—is markedly different from that in the
developing world. Allowing one, but not the other, creates an uneven
playing field. Similarly, the IMF often forces developing countries to
have high interest rates, well above the global ‘market rate’. Lending
money at more reasonable rates should not be viewed as a subsidy.
A third example concerns the sale of privatized assets, particularly
in the former communist economies. Assume that the original
investment was subsidized, but the government privatizes the
industry through a competitive auction. Such an auction should
extinguish the subsidy: the new investor pays, in effect, fair market
value for the asset. In a way one can look at the privatization as a
bankruptcy/restructuring proceeding. When a firm goes bankrupt,
its assets are sold in an auction. The acquiring firm is not viewed as
having received a subsidy. The communist economies can be viewed
as a large bankrupt enterprise, the assets of which are now being dis-
posed of. On the other hand, when the government effectively gives
away the asset, then the subsidy is clearly not extinguished. (A side-
benefit of a rule that distinguishes between the two kinds of privat-
izations is that it would encourage more honest privatizations.)
Summary
Trade negotiations have traditionally focused on expanding market

access. But in the past, the agenda for expanding market access has
been set by the developed countries. More recently, attention has
19
Moreover, dumping is sometimes found in competitive industries, in which no rational firm would ever
engage in predatory pricing, since there is no way it could establish the monopoly power required for it to
recoup the losses it makes when it sells below marginal cost.
focused on the failure to expand market access to agriculture. This

chapter has shown, however, that a true development round market
access agenda goes beyond agriculture, and is markedly different
from the Doha round. It includes unskilled services, migration, indus-
trial tariff structure, and of increasing importance, the non-tariff
barriers that are of increasing importance as tariffs are brought down.
But a true development round should go beyond market access.
There are new issues where international agreements could help
developing countries. We turn to these in the next chapter.
9
Priorities Behind the Border
3
Restrictions on tax and incentive competition

to attract investors
One arena in which an international agreement might be of

immense benefit to developing countries concerns their competi-
tion for investment through concessionary tax rates and financial
subsidies. The main beneficiary of that competition is international
business, and often countries suffer large fiscal losses without com-
mensurate gains to either their domestic economy or to the effi-
ciency of the location of international production.1 If authorities
were to embark on cross-country (or cross-jurisdiction) policy action,
to reduce the harm from such competition to attract business. There
are essentially three options, representing three levels of ambition
with regards to the objectives being pursued. In ascending order
these are: (1) transparency-enhancing obligations on firms and coun-
tries;2 (2) co-operation between jurisdictions;3 and (3) the putting in
place of enforceable international rules.4
1
For a discussion of harmful tax practices see OECD (1998). For welfare losses from international tax
2
competition see Charlton (2003). See Oman (2000).
3
OECD countries adopted a similar approach in their efforts to identify and reduce ‘harmful tax compe-
tition’ (OECD 1998). While the OECD’s mandate here covers mainly general tax rates rather than specific
incentives, the criteria used to determine ‘harmful’ tax policies is instructive for investment incentives. Two
of the criteria cover transparency and discrimination between foreign and domestic firms. The European
Commission’s 1999 ‘Code of Conduct (Business Taxation)’ has taken a similar approach.
4
Three alternative frameworks could regulate incentives with reference to (1) size (capping the total
financial benefit available); (2) use (e.g. specifying geographical areas or sectors in which they are
allowed/prohibited); or (3) instrument (proscribing instruments perceived to be particularly harmful).
Just as international agreements circumscribe subsidies in general,

there should be a strong proscription on firm-specific competition.
The spirit of the WTO’s Agreement on Subsidies and Countervailing
Measures (SCM) could be extended to new rules limiting investment
competition. Under the SCM, subsidies are actionable if they can
be shown to cause adverse trade effects. One of the adverse effects
triggering actionability under Part III is ‘serious prejudice to the
interests of another member’—a principle which could be analo-
gously applied to the incentive instruments used in investment
competition.
The European Union (which has been operating state aid guide-
lines now for several decades) provides an example of how rules
might be developed. Although grants and subsidies to foreign direct
investors are not explicitly targeted by Commission policy, in prac-
tice they are one of the main forms of state aid regulated by it. The
definition of state aid clearly encompasses traditional instruments
of investment attraction. Indeed the European Commission classi-
fies state aid as including (1) grants to firms; (2) loans and guarantees;
(3) tax exemptions; and (4) infrastructure projects benefiting identifi-
able end-users. These payments are regulated by the European
Commission, which claims some success in reducing subsidies in
the EU.5
Anti-corruption policies
While trade in general may benefit developing countries, some kinds

of cross-border transactions clearly harm them. One particularly
insidious interaction between foreign firms and developing coun-
tries is rampant corruption: it is often less expensive to bribe gov-
ernment officials to obtain, say, a concession than to pay the full
market price. International non-bribery legislation (such as America’s
5
See Charlton (2003) for a discussion of the EU’s state aid regulations as applied to foreign investment
incentives.
PRIORITIES BEHIND THE BORDER 135
Foreign Corrupt Practices Act) should be made part of an international

agreement. There should be full disclosure of all payments made to
foreign companies (publish what you pay). There should be an agree-
ment that only disclosed payments will be tax-deductible; but even
stronger enforcement measures should be undertaken. There should
also be a commitment to repatriate funds stolen or otherwise ille-
gally obtained (e.g. through corrupt transactions) from developing
countries. And transactions giving rise to other sources of illicit rev-
enues, particularly those which support armed insurrection, such as
‘conflict diamonds’, should be proscribed.
Secret bank accounts facilitate corruption, by providing a safe
haven for funds stolen from a country. This adversely affects devel-
oping countries to a significant degree. There should be an interna-
tional agreement proscribing bank secrecy (the importance of which
has recently been recognized in the case of terrorism). This too can
easily be enforced. No bank should be allowed to deal with any bank
in a country which does not conform to agreed transparency stan-
dards. It should be possible to sue any country that does not enforce
such a sanction (e.g. under provisions similar to those discussed
above under fair competition).
Finally, as we have noted, arms sales have had a devastating effect
on many of the poorest of the developing countries. The developed
countries are the major source of these arms. Developed countries
must make a commitment to restrict these sales.
Anti-civil-strife policies and

pro-environment policies
Trade agreements have largely been designed to expand the scope of

trade, on the premise that trade is beneficial. Trade policy has become
controversial because there are some notable instances where that
does not seem to be the case. The most obvious are trading in arms,
especially small arms, trafficking in diamonds and other minerals
which help finance the purchase of arms, and the narcotics trade.
It has become well accepted that countries that export drugs have a
responsibility for containing the sale of those drugs. This perspective
has been pushed by the advanced industrial countries; as they have
come to recognize their inability to control consumption and the
demand side, they have put increasing responsibility on the supply
side. The same principle should hold for arms trade—it may be far
easier to control the sale of arms than the purchase.
The recent trade dispute known as the Shrimp–Turtle case6 raised
concerns about the WTO’s role in sustainable development issues.
This dispute arose over US restrictions on imports of shrimps from
countries that did not have conservation programs for migratory
turtles. Each year, thousands of sea turtles are killed in shrimp trawl
nets. To protect these endangered animals, the US passed a law to
prohibit the import of shrimp from nations which do not require
shrimp boats to be equipped with ‘turtle-excluder devices’ (TEDs)—
attachments that enable turtles to exit unharmed from nets. The US
measure was challenged at the WTO by India, Malaysia, Pakistan,
and Thailand. These countries argued that the law was an illegal
restriction on their shrimp exports and thus contravened WTO
obligations. In response, the United States argued that their measure
was covered by Article XX of the GATT, exempting WTO members
from their trade obligations in order to protect human, animal, and
plant life (Article XX(b)) or conserve natural resources (Art. XX(g))
when deemed necessary. In its adjudication, the WTO’s Appellate
Body made clear that the WTO gives countries the right to take
trade action to protect the environment, in particular relating to
human health, endangered species, and exhaustible resources.
It argued that the preamble to the WTO recognized the goal of sus-
tainable development as an objective of the organization and this
made environmental protection a legitimate and important goal
of policy, ranking with protection of international trade as a WTO
objective. The Appellate Body also said that measures to protect
sea turtles would be legitimate under GATT Article XX,7 which
6
Appellate Body Report WT/DS58/AB/R, adopted 6 Nov. 1998; original panel report WT/DS58/R and
Corr. 1, as modified by the Appellate Body Report WT/DS58/AB/R.
7
Article XX provides specific instances in which countries may be excepted from WTO rules. These
include two sets of circumstances for environmental protection.
provides exceptions to the WTO’s trade rules so long as certain basic

criteria such as non-discrimination are met. The US lost the case,
not because it sought to protect the environment but because it
discriminated between WTO members. (The United States was
discriminating by giving Asian countries only four months to
comply with its law, but allowing Caribbean Basin nations three
years.) But the significance of the Shrimp–Turtle rulings is that they
endorse the use of trade policy to enforce environmental standards.
There is, of course, a danger here that the economically powerful
nations, principally the US and the EU, but soon probably China,
will be able to impose their political will upon countries which
are economically dependent on uninterrupted access to these huge
markets. But the WTO’s law is evolving in a responsible way which
defers to multilateral attempts to deal with global environmental
problems.
Several environmental treaties already enshrine the right to use
trade policy to enforce the agreement.8 Where two countries
are party to an environmental agreement the WTO should enable
countries to use trade policy to enforce it, consistent with the
agreement. Where there is a multilateral environmental agree-
ment and a signatory country attempts to take action affecting
the trade of a country that has not signed and is not in compliance,
the WTO should also enable this if the treaty is genuinely multi-
lateral (as in the case of the Kyoto Protocol). Countries that have
signed multilateral agreements to deal cooperatively with global
problems should be able to apply the agreement even to goods
and services from countries that have not. Certainly restrictions
on environmentally unsound techniques of production can have
adverse effects on particular developing countries. The discussion
above suggests that nonetheless there may be a compelling case
for such restrictions. This is especially true for global warming,
which may itself have a particularly adverse effect on some
8
There are more than 200 multilateral environmental agreements. About 20 of these include provisions
that can affect trade: for example they ban trade in certain products, or allow countries to restrict trade in
certain circumstances. Among them are the Montreal Protocol for the protection of the ozone layer, the
Basel Convention on the trade or transportation of hazardous waste across international borders, and the
Convention on International Trade in Endangered Species (CITES).
developing countries, like Bangladesh and many of the countries

in the tropics.
Responding to crises: from

beggar-my-neighbor to help-my-neighbor
In the event of a crisis afflicting one member the WTO should

encourage other countries to take special measures to assist it.
Crises can restrict the availability of short-term trade finance and
hence dramatically reduce trade. Trade finance is particularly impor-
tant in developing countries where exporters may have limited
access to working capital and will often require financing to under-
take production before receiving payment. At its Fifth Ministerial
Meeting in Cancún the General Council of the WTO presented a
report to Ministers stating that, ‘Based mainly on experience gained
in Asia and elsewhere, there is a need to improve the stability and
security of sources of trade finance, especially to help deal with peri-
ods of financial crisis’ (WTO Document WT/WGTDF/2).
Already public institutions, including regional development banks,
have had some success in making trade financing and guarantees
available to emerging economies. Trade finance facilities provided
by regional banks to importers and exporters can include the provi-
sion of working capital loans or overdrafts, issuing-performance, bid,
and advance payment bonds, and extending letters of credit (see
Stephens 1998). For example, in 2000 the Asian Development Bank
made available a US$150 million Political Risk Guarantee Facility
to international banks confirming Pakistani letters of credit. The
facility provided open access to any international bank, covering
only political risks, while leaving commercial risks to the banks. In
1998 the Export–Import Bank of the United States provided short-
term insurance for more than US$1 billion of US export sales to
South Korea (see Auboin and Meier-Ewert 2003).
Many types of export assistance, including export credit insurance
schemes, are subject to binding rules under the WTO’s rules on
subsidies and countervailing measures (SCM). The SCM prohibits

subsidies that are contingent upon export performance—and export
credits, guarantees, and insurance may under certain circumstances
fall within the scope of that prohibition. Under Article 1 of the SCM, a
subsidy is defined as (a) a financial contribution (b) by a government or
public body within the territory of a member (c) which confers a bene-
fit. The principles of the SCM might contradict trade finance assis-
tance because export credits, guarantees, and insurance schemes
involve ‘financial contributions’, and central banks, export credit
agencies, and other government-owned or controlled entities that pro-
vide such schemes likely constitute ‘governments’. Thus as well as
working to develop multilateral mechanisms to improve trade finance
during periods of crisis, the WTO must ensure that the SCM does not
restrict this kind of assistance (at appropriate interest rates) during
periods of crisis by governments or regional development banks.
Since trade will ultimately be the route through which countries in
crisis achieve a balance-of-payments recovery, the WTO should
encourage members to assist countries in crisis by undertaking special
measures to open up their markets. For instance, in the Argentine
crisis, if countries had provided special access to Argentinian beef or
wine, it might have modulated the downturn and facilitated a quicker
restoration of the economy. An international panel within the WTO
should assess whether a crisis which might benefit from special trade-
opening measures exists, and how those ‘help-thy-neighbor’ policies
might be implemented.
One of the original motivations for international trade agreements
was the fear of the kinds of beggar-thy-neighbor policies which
marked the Great Depression. Nonetheless, even within the WTO,
there are provisions (almost never invoked) that allow countries to
take emergency measures. The issue is important because given the
unstable global financial and economic system, country after coun-
try has faced a crisis in recent years. By one reckoning there have
been a hundred crises in the last three decades. Rather than resorting
to beggar-thy-neighbor policies, encouraging countries to return to
protectionist measures in the event of a crisis, it would be far better
to encourage other countries to take special measures to open up
their markets.
Trade implementation and

environment facility
The developing countries are at a marked disadvantage, not only in

negotiating fair trade agreements, but also in implementation. We
noted one aspect of this earlier: their difficulty in mounting chal-
lenges to bio-piracy actions under the TRIPS Agreement.
Some developed and many less developed countries give substantial
subsidies to energy consumption, which has adverse effects on the
global environment. The costs of global warming are likely to be par-
ticularly severe for some developing countries, such as Bangladesh.9
The international community has recognized the need to assist devel-
oping countries in facing the incremental costs associated with imple-
menting environmentally sound technologies, which should include
adjustment assistance to help developing countries bear the costs of
eliminating subsidies to (fossil fuel) energy.
Assitance for implementation
Chapter 13 lays out the broader case for why adjustment costs to a
new trade regime are likely to be greater for developing countries,
and why the international community needs to increase it assis-
tance to developing countries for these purposes.
9
More broadly, because on average developing countries lie in warmer climates, and they are more
heavily dependent on agriculture (which itself is more sensitive to climate) the adverse effects on them will
be larger. There are also likely to be further adverse effects on health.
10
What should not be on
the Agenda?
3
The preceding is a partial list of the items that should have a high pri-
ority in any round of trade negotiations that pretends to call itself a
development round. Many of the items listed have received little or
no attention. Equally remarkable are several of the items (especially
within the so-called Singapore Issues) that were put on the table.
Some of these would almost surely impede development. The fact
that the United States and Europe put such items on the agenda and
continued to push them for so long within the so-called Development
Round is of concern: were they merely bargaining chips? Was there no
real comprehension about what should be meant by a Development
Round?
Intellectual property rights
Recent debates about intellectual property need to be put into context.

Intellectual property provides innovators with temporary monopoly
power. Monopoly power always results in an economic inefficiency.
There is accordingly a high cost of granting even temporary monopoly
power, but the benefit is that by doing so, greater motivation is pro-
vided for inventive activity. The dynamic gains, it is hoped, exceed
the static losses.
Much of the most important innovative activity is outside the

realm of intellectual property. Behind the inventions associated with
atomic energy or lasers were basic discoveries in physics. Behind the
computer were basic discoveries in mathematics. The basic research
which underlies practical innovation in almost all arenas occurs in
universities and government research laboratories, and few of these
discoveries are protected by intellectual property.
In many cases, it is neither desirable nor practical for this to hap-
pen. Often the applications which give market value to the discovery
occur years after the original discovery (beyond the normal patent
life). Ideas give rise to other ideas, and there is no way to ascertain
which ideas proved instrumental in the creation of follow-on ideas.
Most important, it should be recognized that material reward pro-
vides little of the motivation for much of this intellectual activity. To
be sure, it could not occur without financial support. The salaries of
the researchers have to be paid, and if the financial support is woefully
inadequate, many would-be researchers will divert their attention to
other areas. Yet there is little evidence that stronger intellectual prop-
erty protection would generate a greater flow of basic ideas.
Knowledge is a public good, and this is especially true for the fruits
of basic research, which is why governments have an important
responsibility for its support. Intellectual property protection thus
constitutes only a part—and not the most important part—of what
may be called our knowledge and research system. Providing greater
support to this one part of the system may actually harm other parts of
the system and impede the progress of science. Note that the system
under which basic research is conducted is a very open one in which
ideas freely move around, and in which, in fact, scientists put consid-
erable effort into disseminating their ideas and encouraging others to
use them. In many ways, this is the opposite of the premises underly-
ing intellectual property, which seeks to circumscribe the use of
knowledge, limiting it only to those who are willing and able to pay.
Thus, whether within the WTO or through an alternative forum,1
a new intellectual property regime needs to be created which
1
For example the World Intellectual Property Organisation (WIPO) might be an appropriate forum. As we
noted earlier, it is not clear that the WTO is the best forum for the establishment and arbitration of intellec-
tual property rights.
WHAT SHOULD NOT BE ON THE AGENDA? 143
balances more carefully the interests of the users and the producers
of knowledge2 and goes some way toward closing the North–South
‘knowledge gap’.
In some areas weaker patent laws are actually necessary to safe-
guard public health and promote development. The use of compul-
sory licensing and government use of patents could be extended for
these purposes. Indeed many governments in developed countries
already have strong national laws for public use of patents. Under
28 USC Sec 1498, the US government can use patents or authorize
third parties to use patents for virtually any public use, without
negotiation.3 Patent owners have no rights to seek injunctions and
may not seek compensation through tort litigation. Other developed
countries have similarly permissive laws. Before NAFTA, Canada
routinely granted compulsory licenses on pharmaceutical products
for the purpose of reducing health costs through widely available
generic drugs. Canada assigned royalties to the patent holders, usu-
ally of 4 per cent of the generic competitor’s sales price. In sharp con-
trast, and despite the HIV/AIDS public health crisis, no African
country has issued a compulsory license for any medicine.4
This situation is curious because, unlike the spirit of many areas
of the WTO’s intellectual property regime, the TRIPS accord pro-
vides quite liberal powers to governments to authorize third parties
to use patents without the permission of the patent owners. For
example, Article 31(b) allows countries to use or authorize a third
party to use a patent without negotiation or without a license if the
use is for public non-commercial purposes, although the provision
does require that ‘adequate’ compensation be made to the patent
holder (Art. 31(h)).
Love (2001) points out that the existing TRIPS accord permits
countries to create very simple and easily administered systems for
permitting production or import of generic products from the com-
petitive sector. In this particular area the key action required to
promote development is not significant reform of the current WTO
rules, but rather it is providing developing countries with the resources
2
Knowledge is a global public good, and thus it is particularly appropriate that the funding for such global
public goods be provided by those that are most able to pay, i.e. those in the advanced industrial countries.
3 4
For a discussion see Love (2001). See Love (2001).
to develop national systems which take advantage of the current

rules. In particular developing countries should be encouraged to
develop compulsory licensing systems which are simple and not
costly to administer. The median cost of US patent litigation in 1998
was US$1.2 million for each party (Love 2001). As a result of the high
cost of disputes, large patent-holding corporations can use litigation
to tie up any system which enables them to do so. Fortunately,
TRIPS permits countries to administer most aspects of compulsory
licensing through administrative processes (see Art. 31(c), (i)–(k)) and
in several cases does not require governments to grant injunctive
relief to patent holders (Art. 44.2), including when related to public
health.
In other areas, the TRIPS Agreement does require further revi-
sion. In particular, compulsory licensing should be extended beyond
national emergencies to broader ‘refusal to deal’ scenarios in which
developing countries are unable to access products patented by cor-
porations which choose not to serve their market, for example,
because it is too small. The revenues lost to the patent-holders as a
result of such compulsory licensing are likely to be small relative
to the revenues generated by the exercise of monopoly power in
the more advanced industrial countries and therefore are likely to
have a negligible effect on the development of new technologies.
By contrast, the cost to the developing countries of failure to
provide technologies at affordable prices, particularly drugs, is
enormous.5
In addition, Article 40 should extend the right of WTO members
to provide in their national legislation for the prevention of anti-
competitive licensing practices with respect to intellectual property
rights. And pursuant to Article 66.2, new and additional measures
need to be developed to ensure the transfer of technology from devel-
oped countries to least developed countries. The Doha Declaration
recognized the potential for trade agreements to promote the transfer
5
The right of a government to demand compulsory licensing has been recognized even by the United
States, the staunchest defender of intellectual property rights. When it was worried about anthrax, it forced
the compulsory licensing of Ciprio. The only issue is under what conditions such compulsory licensing
should be allowed. A Development Round would have provided answers that more directly address the
concerns of the developing countries.
of technology (para. 37) and proposed the establishment of a working

group to develop recommendations on how progress might be made
in this area. Suggested measures in this area include the establishment,
by developed countries, of specific incentives to encourage their
firms to transfer technology to developing countries.6 However, lit-
tle progress has been made in this area, and developing countries
have been slow to demand effective measures.
The disparity between price and marginal cost of production can
be viewed as a tax used to finance research. Basic principles of equity
question the levying of this tax on some of the poorest people in the
world. If the international community believes that there is a need to
provide greater incentives for research for the development of medi-
cines, then they should do so directly, through funding of research
within either the public or private sector, not by levying a tax on the
poor. One proposal has it that each country should make a contribu-
tion to research whose magnitude would be based on their income
and whose form would be of their own choosing. This contribution,
for instance, could be in the form of direct expenditures on research,
licensing fees, or implicit taxes paid to holders of patents.
There are other issues which affect developing countries’ access to
life-saving medicines at affordable prices. One concern is the ease
with which generic drugs are able to get established, and how
quickly they can enter a market at the expiration of a patent. In some
of its bilateral trade agreements, the United States has been working
to make it more difficult. If there is to be an intellectual property
agreement within a Development Round, it should enshrine prin-
ciples facilitating the rapid entry of generics.
The problems posed by bio-piracy are equally serious. While, as
noted earlier, some of the claims of Western firms may not be sus-
tained when contested in court, it is costly for developing countries
to mount the legal challenge. Article 27.1 (the requirement of uni-
versal novelty as a condition for patentability) should be strength-
ened to protect traditional knowledge. This could be done in part by
amending the TRIPS Agreement to comply with the United Nations
6
See ‘Non-Paper Submitted to the Council for Trade-Related Aspects of Intellectual Property Rights’ by
South Africa (WTO Ref: Job(02)/15).
Convention on Biodiversity (CBD), which was signed by 170 coun-

tries in 1993. The CBD recognizes the collective rights of village
communities over those of individuals or companies and decrees
that a rich country’s demand for patent rights should not come at the
expense of providing incentives for the conservation of plant diver-
sity. One proposal is that there be a change in the presumptions asso-
ciated with patenting, say, traditional medicines, with the applying
party having to show that there has been no previous recognition of
its medicinal properties, with the adjudication occurring in an inter-
national tribunal, and with the legal expenses of the developing
country being divided between the applicant and the developing
country in proportion to the ratio of the income per capita.
The argument of Bhagwati and others that intellectual property
should not be included in a trade agreement is sufficiently com-
pelling that in fact there should be a complete rollback of the TRIPS
Agreement. The issues should be switched to another international
forum (e.g., WIPO). Whether within the WTO or this alternative
forum, a new intellectual property regime needs to be created which
balances more carefully the interests of users in both developed and
less developed countries (including researchers, for whom knowl-
edge is one of the most important inputs) and producers of knowl-
edge. This should be reflected in all the provisions, including the
tests of novelty,7 as well as the breadth and scope of the patent. There
should be a stronger presumption in the case of narrowly defined
patents, and the issue of patents for business practices as well as
other recent extensions of patent coverage should be examined and
agreed to within an international process that is centered in the
scientific community, not the trade ministers. There should also be
sensitivity to the disadvantageous positions of developing countries
in pursuing legal recourse.8
7
Patents could not, for instance, be granted for traditional medicines or goods, or slight variants of those
traditional medicines, when the usefulness of those commodities has already been recognized within the
developing country.
8
There is already in motion a backlash among the more technologically advanced of the less developed
countries. Brazil is pushing for open source software, and China may adopt its own telecommunications
standards which will enable it to avoid paying high royalties for the use of technology based on other stan-
dards. An unbalanced intellectual property regime can contribute to overall global inefficiency in the use and
production of knowledge.
Competition Issues
Competition was supposed to be one of the Singapore Issues, but the
discussions on competition have devolved more toward ensuring
fair competitive access of developed countries to developing country
markets than into ensuring that markets are really competitive, and
that developing countries have fair access to developed country
markets.
Ensuring competition
Today, many companies operate across boundaries. Competition
policy in one country can affect others. The United States and
Europe have increasingly come to recognize this. The United
States instituted an action against Japan, claiming that anti-
competitive practices in Japan (which the Japanese government
had not stopped) had unfairly discriminated against Kodak. Europe
took actions against Honeywell on anti-trust grounds, and the
American government complained that its standards were too
high. The EU is considering taking actions against Microsoft; even
though American courts have found Microsoft guilty of violations
of anti-trust laws, there is widespread concern that the remedies
were insufficient.
The concerns are two-sided: there is a worry that anti-trust laws
will be applied in a discriminatory way, to hurt foreign companies,
and that anti-trust actions will fail to take account of anti-competitive
effects in developing countries. Ideally, there should be harmoniza-
tion of anti-trust laws at the highest standard. Advocates of strong
competition worry, however, that harmonization will occur at the
standard of the least common denominator, and an international
agreement will legitimate such lower standards.
Given these difficulties, initial steps would include insisting on
national (non-discriminatory) treatment. This would entail either
eliminating dumping duties or revising anti-trust legislation, to
apply the same standards to foreign and domestic firms.
A second reform would require that national authorities look

carefully at anti-competitive effects outside their own jurisdiction.9
Not only should domestic anti-trust regulators look at competitive
effects abroad, but foreign consumers should have the right to take
actions in foreign courts against corporations that abuse their mar-
ket power. Cross-border class-action suits should be sanctioned,
allowing consumers in multiple jurisdictions to band together to
impose, for instance, treble damages, with judgments enforceable in
the jurisdiction of the home country.
Third, consumers and governments in all countries should be able
to take actions (including class-action suits) against international
cartels, including those cartels in which governments are a party or
which they have sanctioned. (While some developing countries may
lose from such an action, the benefits received by others would
almost surely outweigh the losses. For instance, oil producers may
be worse off, but oil consumers would be better off.)
Ensuring fair access

With respect to fair access, the concern of developed countries is that
restrictions imposed by developing countries (such as affirmative
action or preferences for small and medium-sized enterprises) have a
differential effect on multinationals. But a Development Round
should recognize the legitimate role of such restrictions as social and
developmental policy tools, and there should be a high burden of
proof imposed on any challenge to such restrictions in order to
establish that the restriction has no legitimate social or develop-
mental objective or that those objectives could not be practicably
achieved10 in a significantly less trade-distorting way. At the same
time, developed country regulations and practices which have an
adverse effect on firms from developing countries (e.g. high licensing
fees) should be held to a much higher standard.
9
Fink, Matoo, and Rathindran (2001) suggest that the GATS should require domestic competition law
to consider the effect of collusive agreements on foreign markets. (The relevance of this point, of course,
goes well beyond the service sector.)
10
For example, without significant adverse effects on other groups. The alternative should be ‘Pareto-
superior’.
Investor agreement
There are two separate issues that need to be considered; one is the
desirability of an investment agreement and the terms which a pro-
development agreement should embrace, and the other is whether
such an agreement should be part of a trade agreement.
On the second issue, we have our doubts, for several reasons. First,
the principle of conservatism articulated in Chapter 5 says that trade
agreements should focus on trade. The Uruguay Round tried to
expand the remit by including both intellectual property and invest-
ment, lightly clothed under the pretense that only trade-related mea-
sures were being considered (thus, we have the TRIPS Agreement,
for trade-related intellectual property, and the TRIMS Agreement, for
trade-related investment measures). But especially under the TRIPS
Agreement, there is little in intellectual property law that is not
trade-related, in the superficial sense that almost all intellectual
property is in some way embodied in tradeable goods.
Second, taking a broader perspective, why were labor and environ-
mental issues not also included? After all, environmental and labor
regulations also affect trade. The refusal of the US to restrict carbon
emissions gives American producers an advantage over European
producers in energy-intensive products. Firms that employ workers
on terms that do not comply with core labor standards may have a
competitive advantage over those who do.
More generally, we have our doubts about the importance to devel-
oping countries of an investment agreement, at least along the lines
conventionally discussed. The absence of a multilateral agreement
has not prevented substantial unilateral liberalization of investment
regimes. UNCTAD (2002) reports that between 1991 and 2001, a total
of 1,393 changes were made to national investment regulations and
more than 90 per cent of these were liberalizing. In 2001, over 200
regulatory changes were made in 71 countries, only 6 per cent of which
were restrictive. In this environment there does not seem to be a com-
pelling rationale to force national governments to adopt a uniform
multilateral agreement. Idiosyncratic national regimes are often more
sensitive to national development priorities than one-size-fits-all
multilateral disciplines.
Moreover, if, as the advocates of these investment measures

claim, they are good for developing countries because they will help
attract investment, countries will have an incentive to introduce
them. Indeed, individually, their incentives to do so may exceed
their collective incentives, because those that do adopt these provi-
sions (according to this theory) would be viewed as more attractive
sites for investment than those that do not.
In fact, the historical experience provides little evidence that
investment treaties generate significantly increased investment
flows. Bilateral investment treaties (BITs) surged in the 1990s to
more than 2,000 in 2001. There was significant activity between
developing countries, which accounted for 42 per cent of new BITs in
2001 (UNCTAD 2002). BITs often proscribe a range of investment
protections that go further than many of the realistic proposals
before the WTO. Yet there is not much evidence that the signing of
bilateral investment treaties increased the flow of investment.
UNCTAD (1998) found no relationship between the level of FDI and
the number of BITs signed by host countries. A more comprehensive
study by Hallward-Driemeier (2002) looked at the bilateral flows of
OECD countries to 31 developing countries over twenty years. After
accounting for trends, they found little evidence that BITs increased
investment to developing countries. More research needs to be done
on the effects of investment treaties on investment volume, but the
existing evidence suggests that the benefits of additional treaties
may be small.
If there is to be an investment agreement (either within the WTO
or not) then the major subject of concern for developing countries is
the race to the bottom—competition for investment that will erode
taxes, environmental standards, or labor conditions. The kinds of
investment agreements that have been pushed within the
Development Round have focused on quite different issues, of bene-
fit to the developed, not the developing, world.
Developed countries have put considerable efforts into expanding
investor rights. As we noted earlier, facilitating the free mobility of
capital is far less important for global economic efficiency or for the
developing countries themselves than facilitating the movement of
labor, particularly that of unskilled workers. Indeed, there is a strong
case that capital market liberalization may actually lower global

economic efficiency.11
Moreover, as we have also noted, to the extent that there is valid-
ity to the argument that improved investor protections will attract
more capital, each country can do that on its own. A developing
country does not have to rely on an international agreement. But it is
only through international negotiations that free labor mobility can
be achieved.12 There is, accordingly, a far stronger argument for
focusing on the ‘rights of labor’ than on the ‘rights of capital’.
Equally troublesome is that arguably, some of the items that are on
the agenda would actually have an adverse effect on the well-being of
developing countries. The United States put the issue of capital mar-
ket liberalization on the table, and has in fact insisted on such provi-
sions in bilateral agreements (e.g. with Chile and Singapore). There
is mounting evidence that full mobility of short-term speculative
capital (‘hot money’) would actually increase economic instability,
in turn increasing poverty. There is little evidence that it enhances
economic growth. Indeed, the instability which it generates may
well impede investment and growth. The problems of Latin America
in recent years, and of East Asia at the end of the last decade, can be
directly traced to capital market liberalization.
The problem with many investor protections is that other rights
have been compromised in the attempt to enhance the rights of
investors. Such investor rights are not costless. But those whose
rights are being compromised do not have a seat at the table (see the
discussion of institutional reforms in Chapter 9). For instance,
Chapter 11 of NAFTA granted investor rights which compromised
the rights of government to provide for the general welfare through
health, safety, and environmental regulations. Recent decisions sug-
gest that the right of a community to protect itself against toxic
wastes may be compromised.
11
See e.g. Stiglitz (2000).
12
The distinction is perhaps not quite as strong as it has sometimes been put. Allowing immigration of
labor will benefit both the recipient and donor country; but there are likely to be groups that are directly
adversely affected in the recipient country and who will be vocal, and often politically effective, in their oppo-
sition. On the other hand, investors seldom oppose capital market liberalization, as they focus on the con-
sequences of the lowering of the cost of capital. Of course, entrenched industry may resist the entry of
competitors in their line of business.
There are already mechanisms for the protection of investors

against expropriation, both internationally (e.g. MIGA, the Multi-
lateral Investment Guarantee Agency) and on the part of many
of the investing countries (e.g. the Overseas Private Investment
Corporation, OPIC, in the United States). A convincing case has not
been made that these are inadequate, or, if they are, that they cannot
be strengthened. The new investor protections go beyond the
concern for expropriation, to the granting of additional rights to
investors.
Other services
In our list of priorities, we emphasized earlier the opening up of mar-

kets to unskilled labor-intensive services and the movement of
unskilled labor (sometimes in support of such services). Earlier
rounds of trade liberalization focused on, for instance, financial ser-
vices, the benefits of which are arguable. The standard argument is
that more efficient financial service intermediation lowers the cost
of doing business and thus promotes economic growth. It is pro-
development. But a closer look at the record reveals a more mixed
picture. In at least some developing countries there are concerns that
the purchase of local banks by foreign banks has reduced the flow of
credit to small and medium-sized domestic enterprises, and thus
impeded economic growth. (There is a long history of such concerns,
evidenced in the United States for instance by restrictions on inter-
state banking, intended to prevent New York and other money-
center banks from buying up other banks, thereby impeding regional,
and especially rural, development.) Agreements on financial ser-
vices should be re-examined to ascertain whether there is sufficient
protection for developing countries. In particular, the right of devel-
oping countries to impose lending requirements to force more lend-
ing to underserved populations (analogous to those in the United
States in the Community Reinvestment Act) should be explicitly
recognized.
Other regulatory interventions

Developing countries worry that new trade agreements will create
new barriers to the entry of their goods into developed country mar-
kets (impeding their development). They worry about blue tariffs
(impediments based on labor standards) and green tariffs (impedi-
ments based on environmental standards).
Standard economic theory suggests that, with a couple of excep-
tions noted below, weak standards do not necessarily improve a coun-
try’s competitiveness, and therefore the issue of standards should
not, in general, be embraced within a trade agreement. In standard
theory, in a competitive market, any costly provision (such as
improved working conditions) simply gets reflected in the wage paid.
Such restrictions affect the form of compensation, but not the overall
level of compensation. In general, there is no reason for the interna-
tional community to intrude into the forms of compensation.
There are three basic exceptions to these principles. The first is
when the global community is affected (a principle which has already
been recognized in the appellate decision in the Shrimp–Turtle case,
in the area of environment and endangered species13). The interna-
tional community has a right to take actions to address global public
goods and externalities, and among the most important of these is the
global environment. Trade policy should recognize, as we have noted
earlier, that not forcing firms to pay the true social costs of their envi-
ronmental damage is a form of subsidy which countries should have
the right to take action against. Since developing countries as a whole
are more likely to be adversely affected by global warming than, say,
the United States,14 using trade policy to force compliance by the
advanced industrial countries with the Kyoto Protocol could well be
considered an important part of a pro-development trade agenda.
The second is related to the first—matters of human rights.
Clearly, when individuals are forced to provide labor services
13
As noted earlier, the United States requires domestic shrimpers to use protective technology called
turtle-excluder devices, which are a kind of trap door by which turtles can escape from shrimp nets. In 1989,
Congress essentially banned importation of shrimp caught by foreign shrimpers who do not use turtle-
excluder devices.
14
This is because the developing countries, on average, are already in warmer climates.
(e.g. when they are prisoners) or allowed to use child labor, costs of
production may be lowered. As a global community, we do not want
to provide economic incentives for such behavior. On the contrary,
we want to discourage it. By the same token, when governments have
seized land of indigenous peoples, and provided the fruits of that land
to others at discounted prices (even if those prices are above its cost
of acquisition), then that should be viewed as an unfair subsidy.
Countervailing duties against minerals and lumber produced in
many countries would be justified by such a provision.
The third, which may also be related to the first, concerns circum-
stances in which countries can take actions which unfairly affect
costs of production. The most notable example of this is restrictions
on collective bargaining and the right to take collective action. Then,
bargaining relationship between workers and firms is one-sided, and
firms can use their economic power to drive down wages and labor
costs, making their products more competitive than they otherwise
would be.
In all of these cases, some argue that since these are not matters of
trade (though the first clearly constitutes a trade-distorting subsidy)
it is preferable to address these problems through other channels.
Without prejudging the validity of this argument, the fact of the
matter is that there are few other channels. Today, in the absence of
alternatives, trade sanctions are one of the few ways that the inter-
national community can enforce its will, and though resort to such
measures should be carefully circumscribed, the instances enumer-
ated are among those in which sanctions may arguably be justified.15
On the other hand, there are a host of other regulatory interven-
tions which may adversely affect foreign businesses, sometimes dif-
ferentially so, but whose primary motivation is to enhance
economic development. We referred to one earlier—restrictions on
banks that require that they lend certain minimal amounts to small
and medium-sized domestic enterprises and to other under-served
communities. It is a legitimate role for government to undertake
such actions. The United States, Japan, and many other countries did
15
It is important that the decision about whether a trade sanction is to be imposed be taken by the inter-
national community; otherwise special interests within a country may well try to disguise protectionism
behind a cloak of environmentalism or labor rights.
so in their earlier stages of development—and continue to do so.

Because foreign banks may not be in a position to screen such loan
applicants as well as domestic banks can, such regulations may have
a differentially adverse effect on foreign banks.
By the same token, governments may decide that affirmative
action programs are desirable for social purposes, and require that all
employers hire workers from certain disadvantaged groups. These
restrictions might, conceivably, impose greater costs on foreign
firms, who are used to hiring Western-educated individuals, but they
reflect a legitimate aspiration of governments to create a more equal
society.
Exchange rate manipulation
The United States has recently leveled charges of exchange rate

manipulation against China. Global financial markets have exhib-
ited enormous instability. Volatility of exchange rates presents a par-
ticular problem for developing countries. Markets are thin, and thus
subject to both more volatility and manipulability. Government
intervention is, accordingly, often viewed to be desirable. There are
a variety of mechanisms by which the government can affect the
exchange rate, and there are a variety of government policies which
affect the exchange rate indirectly. Bad economic policies (for
instance, large deficits) may lead to a devaluation of the currency,
whether that is the intent of the policy or not. Given the sizeable
adverse consequences of trade deficits, there should be a presump-
tion that countries which have only a moderate trade surplus are not
engaged in exchange rate manipulation. The complexities involved
suggest that there should be a high threshold test for taking action in
the event of an accusation of exchange rate manipulation, and that,
at the very least, only multilateral trade surpluses, not bilateral trade
deficits, should be presented as evidence of such manipulation.
11
Joining the Trading System
3
Accession
Twenty countries have been added to the WTO since its creation in
1995.1 Another 24 countries are negotiating membership,2 of which
eight are least developed countries (LDCs). Most of this book has
been concerned with ensuring that the rules and procedures of the
WTO deliver benefits to the developing countries within the WTO
system. For the (equally needy) developing countries who are not yet
members, the benefits of multilateral trade reform in the Doha
Round are dependent on the speed and ease with which they are able
to join the WTO, and the terms of their accession. The recent expe-
riences of several acceding developing countries are reviewed below.3
It shows that the WTO’s accession process requires reform in several
respects. In particular, there are no good arguments for maintaining
the double standard which treats countries differently depending on
whether they are existing members of the WTO or not. There is a
need for transparent and objective rules and procedures for accession
negotiations which ensure that the accession process is not unduly
1
For analysis of the experience of accession see Kennett, Evenett, and Gage (2004). The 20 countries
are: Albania, Armenia, Bulgaria, Cambodia, China, Croatia, Ecuador, Estonia, Georgia, Jordan, Kyrgyz
Republic, Latvia, Lithuania, Macedonia, Moldova, Mongolia, Nepal, Oman, Panama, and Chinese Taipei.
2
In order from oldest to most recent application: Algeria, Russian Federation, Saudi Arabia, Belarus,
Ukraine, Sudan, Uzbekistan, Vietnam, Seychelles, Tonga, Kazakhstan, Azerbaijan, Andorra, Laos, Samoa,
Lebanese Republic, Bosnia and Herzegovina, Bhutan, Cape Verde, Yemen, Serbia and Montenegro,
Bahamas, Tajikistan, Ethiopia, and Libya.
3
For reviews of the experience of individual countries see Grynberg and Joy (2000): Vanuatu; Oxfam
(2003a): Cambodia; and Oxfam (2004): Vietnam.
costly for developing countries and that accession terms reflect the
level of development of each new member.
The WTO’s accession process is lengthy and arduous. At the Fifth
Ministerial Conference, held in Cancún in September 2003,
Cambodia and Nepal became the first two least developed countries
to gain membership through the accession process. The times
between their initial application and approval were nine years and
fourteen years respectively. Vanuatu, Sudan, Samoa, and Laos are all
currently in protracted accession negotiations which began more
than six years ago.
Indeed, the time taken to complete the WTO’s accession process
appears to be growing. Figure 11.1 plots in sequential order the
length of time taken to complete the first twenty accessions. Even
without the longest accession (China, the fifteenth nation to join the
WTO, in 1995), the trend is rising (Kennett, Evenett, and Gage 2004).
Part of the problem lies in the absence of clear and transparent
accession criteria and negotiation guidelines. The formal rules of
accession are set by Article XII of the Marrakesh Agreement estab-
lishing the World Trade Organization. It simply states that new mem-
bers may accede on ‘terms to be agreed between it and the WTO’
(Art. XII.1) and that the Ministerial Conference shall approve the
200
180
160
140
Number of months
120
100
80
60
40
20
0
1 3 5 7 9 11 13 15 17 19
Figure 11.1. Length of accession process for the first 20 countries to join
the WTO
Source: Kennett, Evenett, and Gage (2004).
JOINING THE TRADING SYSTEM 159
agreement on the terms of accession by a two-thirds majority of the

members of the WTO (Art. XII.2). Unfortunately, Article XII does
not give any guidance on how those terms should be formulated, nor
does it provide any membership criteria, or any elements in an acces-
sion process. Indeed for an organization which prides itself on being
‘rules-based’, the accession process is remarkably vague. This lacuna
has left the accession process in the hands of member countries who
have interpreted the ‘terms to be agreed’ as an opportunity to go
beyond the existing WTO agreements and impose extra conditions
(often referred to as ‘WTO-plus conditions’) on acceding countries.
The US has been particularly successful in seeking WTO-plus provi-
sions from its partners in bilateral trade agreements, and then using
these as leverage in WTO negotiations. For example, Vietnam agreed
to TRIPS-plus commitments in its bilateral trade agreement with
the US. Now that Vietnam is seeking accession to the WTO, which
operates under the Most Favoured Nation (MFN) principle, these
concessions have become the effective starting point for accession
negotiations.
In practice the accession process occurs in three stages. In the first
stage the WTO establishes a working party, and the acceding coun-
try submits a memorandum explaining the detail of its trade regime.
In the second stage the working-party members focus on identifying
where the applicant’s trade policies differ from the rules of the WTO.
In the third stage the applicant undertakes a series of bilateral nego-
tiations with working-party members which culminates in an acces-
sion agreement for approval by the Ministerial Conference. Since
developed countries are always represented on the working party
and since all members of the working party must agree to the terms
of accession in order for it to be approved, the developed countries
have an effective veto over the accession of any country that does not
accept their terms. Samoa’s senior trade consultant complained
about the brinkmanship practiced by developed countries: ‘They can
ask for all sorts of commitments which Samoa isn’t in a position to
offer. If they insist, there are two options: we will never become a
member or we have to give in to that request.’4
4
In an interview with Tuala Falani Chan Tung published in Sunline, Apr. 2004.
Recognizing the weaknesses of the current arrangement, the

WTO’s members committed themselves in the Doha Declaration ‘to
accelerating the accession of least-developed countries’ (para. 9).
Following this decision, at the General Council in December 2002,
the WTO membership adopted guidelines for LDCs accession.
According to the guidelines, negotiations for the accession of LDCs
will be facilitated and accelerated through simplified and stream-
lined accession procedures. In addition the guidelines suggested that
members should ensure that acceding LDCs are not subjected to
obligations or commitments that go beyond what is applicable to the
existing LDC members.5 Yet, in several areas, the WTO’s actions fall
short of this rhetoric. The guidelines are not specific—there are still
no objective membership criteria—and evidence from countries in
the accession process indicates that they continue to be treated more
harshly than existing members.
Recent-accession countries have been subjected to a glaring double
standard in that they have been required to accept a higher level of
market access commitments than existing members in many areas of
the agenda. For example, Cambodia, Nepal and Vanuatu have had to
commit themselves to more comprehensive tariff bindings and lower
levels of tariff peaks. By way of comparison, while Cambodia, Nepal,
and Vanuatu have agreed to bind 100 per cent of their tariff lines, other
countries that are already members of the WTO have often bound a
much smaller share of their tariff lines.6 Furthermore, while
Cambodia’s bound rates are as high as 60 per cent for sensitive agri-
cultural products, many least developed (and even developed) WTO
members have much higher bound rates on agricultural goods.7 In
industrial goods, Cambodia was required to bind its tariff rates at a
5
A decision of the General Council on Streamlining Accession of the LDCs, dated 10 Dec. 2002 states:
‘WTO Members shall exercise restraint in seeking concessions and commitments on trade in goods and ser-
vices from acceding LDCs, taking into account the levels of concessions and commitments undertaken by
existing WTO LDC Members’; and ‘Acceding LDCs shall offer access through reasonable concessions and
commitments on trade in goods and services commensurate with their individual development, financial
and trade needs, in line with Article XXXVI.8 of GATT 1994, Article 15 of the Agreement on Agriculture, and
Articles IV and XIX of the General Agreement on Trade in Services’.
6
See UNCTAD (2004). For example, both Tanzania and Cameroon have bound 13.3%, and one devel-
oped WTO member, Australia, has bound 97%.
7
For example, bound rates are as high as 550% in Myanmar, and other developing WTO member countries
have bound tariff rates on agricultural goods as high as 3,000% (Egypt), while developed WTO members have
bound tariff rates on some agricultural goods as high as 350% (United States). See UNCTAD (2004).
maximum level of 50 per cent, much lower than other least developed
countries.8 Similarly, in its accession negotiations, China had to agree
to large tariff reductions in agriculture which went far beyond the
obligations of existing members. It also agreed to a special safeguard
clause allowing individual WTO members to take measures to limit
imports of Chinese products in case of a surge. This extraordinary
measure, which is in breach of the MFN principle and goes far beyond
the GATT safeguard provisions (Art. XIX), could open the floodgate for
the application of discriminatory measures against China. In services,
Nepal was required to open up seventy service sub-sectors—far more
than similar developing countries. Similarly Cambodia was asked to
undertake a commitment on audio-visual and distribution services—
an area on which no existing LDC member has made a commitment
(Oxfam 2003a). It seems strange that the WTO’s developed country
members should force acceding countries, particularly small and poor
countries like Cambodia and Nepal, into such strong concessions.
Grynberg and Joy (2000) suggest that the motivation lies in the devel-
oped countries’ desire to create a precedent that can be applied to
future negotiations.
Moreover there is clear evidence that the price of accession is
growing over time (Kennett, Evenett, and Gage 2004). Figure 11.2
60
Average tariff binding (%)
50
40
30
20
10
0
1 3 5 7 9 11 13 15 17 19
Figure 11.2. Average tariff binding on agricultural products permitted to the

first 20 countries to join the WTO
Source: Kennett, Evenett, and Gage (2004).
8
Other least developed countries also have relatively high tariff bindings in the non-agricultural goods
sector. Bangladesh, Djibouti, Lesotho, and Niger all have peaks in tariff bindings at levels as high as 200%.
See UNCTAD (2004). Nepal has bound tariff rates on non-agricultural goods at a maximum level of 130%.
shows that the average tariff binding on agricultural goods allowed

to acceding countries is falling over time, with the exception of the
last two countries, which are in a different category because they are
the only two LDCs that have joined the WTO. The figure demon-
strates that acceeding countries are making larger commitments on
their agricultural tariffs.
The terms of accession agreements should reflect the level of
development of the new member, yet much recent experience sug-
gests that acceding countries have been forced to offer concessions
that have not been made even by developed countries. For example,
Cambodia and Vanuatu made a commitment to bind export subsid-
ies in agriculture at zero and not to apply such subsidies in the
future. This means that these two developing countries have effect-
ively forgone the right to use export subsidies for agricultural goods,
a right that is granted to all other least developed member countries
by the Agreement on Agriculture. This commitment goes further
than most developed countries have been willing to go in restricting
their own export subsidies.
As well as making more commitments than existing members,
acceding developing countries have been denied access to different
types of special and differential treatment. Cambodia, Nepal, and
Vanuatu accepted a significant reduction of their rights to special
and differential treatment—rights that are granted automatically
to other least developed countries that are already members of the
WTO. New members have been granted much shorter transition
periods than existing members. Cambodia obtained the longest
transition period—five years for customs valuation, three years
for technical barriers to trade (TBT), four years for sanitary and
phytosanitary measures (SPS), and three years for TRIPS. Nepal
negotiated a three-year period for customs valuation, TBT and
SPS, and TRIPS. Vanuatu received only a one-year transition period
for customs valuation and a two-year period for TRIPS (UNCTAD
2004).
Insensitivity to the development needs of new members has been
particularly manifest in the area of intellectual property rights.
Developing countries have been reluctant to commit themselves
rapidly to protecting the intellectual property of firms from the

industrialized countries because that might hinder technology
transfer, enable those firms to extract monopoly rents, and limit
access to crucial technologies such as medicines to combat disease.
For these reasons developing countries have sought long imple-
mentation periods for intellectual property commitments. The
original TRIPS Agreement, which came into effect on 1 January
1995, gave developing countries an eleven-year transitional period
to bring their national legislation in line with the agreements.
Additionally, the Doha Declaration on the TRIPS Agreement and
Public Health increased the transitional period for pharmaceutical
products to 2016.9 In their accession negotiations both Cambodia
and Nepal were asked to commit themselves to elements of the
TRIPS Agreement earlier than other least developed countries.
Cambodia and Nepal obtained a three-year transition period, while
Vanuatu obtained two years. Cambodia made explicit commit-
ments to comply with all obligations concerning patents and pro-
tection of undisclosed information (Part II, sects 5 and 7 of the
TRIPS Agreement), although the Doha Declaration on the TRIPS
Agreement and Public Health provided an exemption until 2016
from these commitments. In other words, Cambodia has been
required to eliminate immediately the use of affordable new
generic drugs even though existing LDC members of the WTO need
not do so until 2016.
There is an urgent need for the Doha Round to reform its accession
procedures in accord with the General Council decision of
December 2002, described above. To make these principles opera-
tional, there is a need to establish clear and objective rules and disci-
plines for accession negotiations. Members should ensure that the
accession process is not excessively costly for the LDCs and that
technical assistance is provided to assist them. It is also necessary
that the terms of accession agreements are tailored to LDCs’ levels
of development and do not force commitments on acceding coun-
tries that are not required of existing members.
9
The Declaration on the TRIPS Agreement and Public Health, WT/MIN(01)/DEC/W/2, 14 Nov. 2001.
Regional and Bilateral trade agreements and

South–South trade
With the collapse of the Doha Round, the United States threatened
to pursue a set of bilateral and regional trade agreements. Such agree-
ments are against the spirit of the multilateral trading system, which
has been based on most-favored-nation principles. Moreover, devel-
oping countries may be even more disadvantaged in one-on-one bar-
gaining with the United States; a series of such agreements may
leave many developing countries worse off than they would be even
with another unfair multilateral agreement. While in principle, such
agreements are only consistent with WTO rules if there is limited
trade diversion, in practice, little attention has been paid to this
requirement. The argument that these regional agreements are use-
ful as a step towards improved multilateral agreements is also sus-
pect. The kind of multilateral agreement to which they may lead
may be more unbalanced than one which would be directly entered
into, without the more circuitous route. More to the point, there is a
high cost to the roundabout approach. For to the extent that there is
temporary trade diversion, some industries are being temporarily
encouraged, only to be later discouraged. Adjustment costs are typi-
cally high in developing countries; there may be significant costs of
entry and exit, and with a scarcity of capital, the burden on develop-
ing countries may be particularly large.
In the next chapter, we propose institutional reforms that address
these issues. But there is another arena in which trade agreements
which fall short of full multinational agreements may be desirable,
and those entail reducing the restrictions on South–South trade.
South–South trade makes up some 40 per cent of developing country
trade. And, as we noted earlier, developing countries have much
higher tariff levels than developed countries. This suggests that con-
siderable gains are available from liberalization of South–South
trade.
How then should WTO agreements manage the trade-off between
the need to protect domestic markets from competition from devel-
oped countries, while not hindering trade between developing
countries? One solution might be to allow developing countries to

offer preferential market access deals to each other in the same way
that the EU’s Everything but Arms initiative offers free access to
developing countries only. Such a provision would have one further
advantage: preferential treatment by some developed countries
(most notably the US) is based on political conditionality—it is a
‘gift’ that is constantly in danger of being taken away, and this
induced risk limits the value of the gift. Moreover, the granting of
such preferences becomes a lever with which developed countries
can extract other concessions from developing countries. A
South–South agreement would likely focus more exclusively on
trade issues, and would not suffer from these abuses.
12
Institutional Reforms
3
Procedures
There is widespread dissatisfaction with the way that trade agree-

ments are made, partly stemming from a belief that current proce-
dures put developing countries at a marked disadvantage. This is
particularly important, given the increasing role that such trade
agreements play in our societies. They define a broad set of rights and
obligations, yet they are arrived at in a manner that is distinctly dif-
ferent from the way that other kinds of legislation are adopted. The
terms are often negotiated behind closed doors, with little public
debate about specific provisions. The legislative process is often
truncated. The result is agreements, like Chapter 11 of the NAFTA
agreement or the TRIPS Agreement, which contain provisions that
would probably never have been accepted by a democratic parlia-
ment with open discussion in a deliberative process.
The hallmark of earlier trade agreements is that they were con-
ducted in secret, with many of the terms not fully disclosed until the
end of the negotiations, and then governments faced an ‘all or noth-
ing’ choice. Because parliamentarians could have no effect on the
outcome, they had little incentive to understand the intricacies.
Given the extent to which trade issues overlap with other issues,
including those touching on the environment, it is important that
the procedural reforms make deliberations about trade issues more
like other deliberative processes—including more open deliberation.
Trade has become too important to be left to trade ministers alone.

Part of the deliberative process must entail the active involvement
of others (meeting together, not just through the trade ministers).
Thus when intellectual property matters are being discussed (if they
are to be discussed at all), the science ministers must be involved.
When trade policy affects the environment, there must be some
mechanisms for the environmental ministers’ voices to be heard.
They would insist, for instance, that provisions be inserted that pre-
vent a race to the bottom, that low environmental standards (e.g.
those associated with allowing the pollution of the world’s atmos-
phere) be viewed as a form of subsidy to be prohibited.
We emphasized in Chapter 5 that the fairness of the international
regime is to be judged not only in terms of the outcomes, but also in
terms of the processes. There is now a large literature documenting
the deficiencies in the procedures,1 and for reasons already noted,
these procedural deficiencies disadvantage developing countries.
That is why procedural reforms—particularly those relating to
transparency and representation—should have a high priority. Civil
society should be given a greater role in the negotiations.
The developed countries should continue the kind of support they
have provided to help the developing countries participate more
effectively in these deliberations. Trade negotiations involve com-
plex issues, including economic issues on which even the experts are
not in agreement. Meaningful participation in these discussions
requires understanding these complexities, knowing the full import
of each of the provisions, and how they might affect countries differ-
ently in quite different situations.
Structures and representation
As the number of WTO members has grown, and the demands for a
more inclusive bargaining process have increased, the current sys-
tem appears to be increasingly unwieldy. It is not the intent of this
1
On participation see Blackhurst, Lyakurwa, and Oyejide (2000); on transparency see Francois (2001);
on representation see Winters (2001).
INSTITUTIONAL REFORMS 169
chapter to provide a detailed analysis of alternative proposals for

institutional reform, but rather to highlight its importance and to
emphasize why such reforms should in fact be viewed as a priority in
current discussions.
It is apparent that the opening up of the WTO to so many members
has made negotiations cumbersome and difficult. This has been used
as an excuse for secretive processes in which, somehow, a selected
group of countries are chosen to negotiate with the United States,
the EU, and Japan in the ‘Green Room’. The Green Room process has
been formally abolished, but small negotiating groups continue to
set the agenda behind closed doors with inadequate openness and
transparency. In other areas of democratic decision-making, espe-
cially those based on consensual processes, as trade negotiations are
in principle supposed to be, the principle of representativeness is
well accepted: a small group of countries is chosen to reflect the var-
ious interests and constituencies—say the largest trading countries,
the United States, EU, Japan, China; a representative or two of the
middle-income countries, say Brazil and one other country; a couple
of representatives of the least developed countries; a representative
of the Cairns Group, etc. Each would then consult with those they
are representing on a regular basis. An open and transparent process
would ensure that the views and voices of all are heard.
Trade negotiations entail a myriad of proposals for changing the
rules of the game, and developing countries are often at a disadvan-
tage in assessing the impact of each of these proposals on them-
selves, let alone the general equilibrium impact on the global trading
system. Another requirement for a fair trade regime would be a new
body within the WTO responsible for assessing the impacts of pro-
posed trade provisions on development and developing countries. Its
mandate would be to look objectively at the consequences of alter-
native proposals for all the countries of the world, recognizing that
economic science is not at a stage where there is agreement about
the ‘right’ model. Thus, such a body might attempt to assess the
impact of the allegedly non-trade-distorting agricultural subsidies in
a world in which there are capital constraints.
Such a body might help too in the enforcement of current agree-
ments, providing guidance, for instance, on whether a particular
proposed bilateral or regional trade agreement is consistent with the

principle that ‘trade diversion’ should be limited, and should be less
than the amount of trade creation.
Other functions of an expanded WTO secretariat might include an
independent body to assess countries in crisis, to adjudicate and
approve the imposition of trade restrictions (‘safeguard measures’),
and to investigate dumping charges, countervailing duties, and
phytosanitary conditions.
There is a need to address the scope of technical assistance and the
capacity of the WTO to provide this adequately within existing
structures. Helping low-income countries strengthen their institu-
tional capacity to permit them to meet WTO agreements will
require not only technical assistance but also significant financial
assistance. The costs of implementation of WTO commitments are
very substantial.2
While a limited assistance program may assist developing coun-
tries to implement reform, technical assistance is not sufficient to
deal with the economy-wide adjustment costs associated with struc-
tural change. These costs, which generate domestic opposition to
trade liberalization, are no less important barriers to progress than
the lack of institutional capacity.
In addition, lack of institutional capacity has the effect of limiting
the access of developing countries to justice within the dispute sys-
tem. Developing countries are disadvantaged in complex and expen-
sive legal proceedings. An expansion of existing legal assistance
schemes will be an important prerequisite for institutional fairness.
Responsibility for the bulk of technical assistance has fallen on
international organizations. Both the World Bank and the WTO have
increased their technical cooperation activities. However, as much as
90 per cent of financing for these activities is provided by trust funds
provided by two or three bilateral donors, while the WTO itself has typ-
ically allocated less than one per cent of its total annual budget for tech-
nical cooperation activities—less than half a million US dollars
(Michalopoulos 2000).
2
Finger and Schuler (1999) ‘The implementation of Uruguay Round Commitments: The development
challenge’, Policy Research Working Paper no. 2000, World Bank, Washington, DC.
13
Trade Liberalization and
the Costs of Adjustment
3
Introduction
Trade liberalization creates adjustment costs as resources are moved

from one sector to another in the process of reform. When tariffs are
reduced, import-competing firms may reduce their production in
the face of new competition, causing some of their workers and
capital to lie idle for a period. The firm’s laid-off workers will incur
costs while searching for new jobs and may need to invest in retrain-
ing. Governments will be called upon to provide assistance to the
unemployed, while also incurring costs associated with implement-
ing the new systems for managing reform.
To take advantage of the opportunities offered by improved access
to foreign markets, developing countries will be required to make
investments—in infrastructure by government and in new facilities
or technologies by exporters—before they can capitalize on the
opportunities offered by improved access to foreign markets.
Significant trade liberalization will also affect the distribution of
income among factors of production: the relative price of the factor
which is in relative scarcity will decline, while that of the abundant
factor will increase.1 Agricultural subsidies get capitalized in the
1
This is the implication of the renowned Stolper and Samuelson (1941) theorem; but even if the
restrictive conditions under which it holds are not satisfied, there is a presumption that relative rewards to
different factors will change in the way indicated.
price of land, and landowners will lose substantial amounts when

such subsidies are eliminated.2 Because there are large distortionary
costs associated with taxation, there are large societal costs associ-
ated with the compensations designed to mitigate these effects.3
Given the severe constraints on raising taxes in developing coun-
tries, the opportunity cost of funds diverted for even partial
compensation may be very high.
Trade liberalization also reduces tariff revenues; as alternative
sources of revenue are limited the costs of the revenue loss is high.
Thus, either public expenditures get reduced or other taxes are
increased, and either of these may have significant adverse effects on
growth.4
Trade liberalization may impose further costs: the movement
from quotas to tariffs, whatever its merits, may expose countries to
additional risks.5 Developing countries with weak social safety nets
will have to devote more resources to strengthening these safety nets
and will have to mitigate the cost of risks. This too needs to be
viewed as part of the costs of trade liberalization.
In one sense, these adjustment costs can be thought of as the
‘price’ to be paid for the benefits of multilateral tariff reduction.
Together these adjustment costs and trade benefits determine the
net effect of trade reform for each country. The Doha Round has
placed renewed emphasis on the importance of sharing the benefits
of trade reform fairly among developed and developing countries.
However, less attention has been paid to the distribution of adjust-
ment costs among countries.
An understanding of the costs of trade reform is important for at
least two reasons. First, if the ‘development focus’ of the Doha
2
The numbers can be large. A US$4bn annual cotton subsidy, if fully captalized in land values, translates
at a 5% interest rate into US$80bn.
3
Thus, even if the dollar value of the gains to the winners from liberalization are greater than the dollar
value of the losses to the losers, trade liberalization may not be welfare-enhancing when the costs of
compensation are taken into account.
4
Many countries have shifted to greater reliance on the value-added tax, but as Stiglitz (2003) has
argued, this switch may have adverse effects on development.
5
That is, countries now are more subject to the vagaries of international prices. See Dasgupta and Stiglitz
(1977). More generally, trade liberalization may make countries more vulnerable to external shocks, and for
countries in which trade looms large in GDP, the result may be greater macro-economic volatility. See
Easterly, Islam, and Stiglitz (2001), and especially the fuller text of the Michael Bruno Memorial Lecture, ‘Is
there a Workable Macroeconomic Paradigm for LDCs?’, presented by J. E. Stiglitz at the 12th World Congress
of the IEA, Buenos Aires, 27 Aug. 1999.
TRADE LIBERALIZATION AND COSTS OF ADJUSTMENT 173
Round is to have any meaning, then WTO members must be mindful

of the fact that the cost of adjusting to their agreements will have
serious consequences for development. Not only do adjustment
costs fall particularly harshly on the poorest people in the world
because they are least able to afford them, but also the costs consume
resources that would otherwise be spent on alternative development
priorities. For many people, the impact of trade reform will over-
whelm the effects of other economic development programs.
The second motivation for understanding adjustment costs is the
pragmatic need to win political support for reform. High adjustment
costs give some groups a vested interest in the status quo. Identifying
and compensating those groups may be an effective way of removing
impediments to welfare-improving policy changes.
This chapter examines the effect of several sources of adjustment
costs. A theme that runs through the empirical evidence is that the
adjustment process resulting from the proposals emerging from the
Doha Round will have a particularly harsh impact on the people and
governments of developing countries—especially small developing
countries. There are several reasons for this asymmetry. First, devel-
oping countries are most vulnerable to policy shocks because their
export industries are the least diversified—many are dependent on
the export and hence world price of just one or two commodities.
Second, developing countries are likely to have to make the largest
changes to comply with international regulations such as those
embodied in the Singapore Issues. Third, the structure of world trade
is most distorted in the industries of importance to developing coun-
tries. World markets for agriculture, processed foods, textiles, and
other critical goods are those most distorted by developed countries’
tariff policies. Consequently these industries will be highly affected
by liberalization—even where reform has long-run net positive
effects for developing countries, they will have to cope with adjust-
ment costs, investment costs, and redistributive effects. Fourth, and
most importantly, developing countries are home to the world’s
poorest people and the weakest credit markets. These people are
particularly vulnerable to adjustment costs.
For these reasons, the adjustment to new trading rules is a radic-
ally different experience for developed and developing countries.
This chapter studies the process of adjustment and the costs it

implies for developing countries. It notes that there are several
policy measures that should accompany trade reform to minimize
the costs of adjustment and the disproportionate welfare losses to
particular social groups. Policies to assist developing countries to
benefit from reforms are also surveyed briefly. Many of these policies
will require assistance from developing countries and international
institutions.
Costs of adjustment
Empirical studies have attempted to define and quantify adjustment

costs in developed countries.6 These studies suggest that labor bears
the brunt of the costs, but that (ignoring the distributional con-
sequences) for developed countries, the costs are small relative to the
gains. Baldwin, Mutti, and Richardson (1980) analyse the adjust-
ment costs for the US resulting from a 50 per cent cut in domestic
tariffs. They focus on adjustment of the capital stock and the costs
borne by laid-off workers. They find that labor bears almost 90 per cent
of the adjustment costs. In total they conclude that adjustment costs
account for 4 per cent of the gains from liberalization. De Melo and
Tarr (1990) use a computable general equilibrium approach to
analyse the welfare effects and adjustment costs resulting from the
elimination of quotas in textiles, steel, and autos. In their model
adjustment costs are measured as the lost earnings suffered by
dislocated workers. They estimate that the adjustment costs are just
1.5 per cent of the gains from liberalization. Winters and Takacs
(1991) found that the British footwear industry suffered just over
1,000 lost jobs as a result of the removal of quotas in the late 1970s
and most of these workers remained unemployed for between 5 and
21 weeks. The study concludes that the lost income in the first year
after the quotas were eliminated amounted only to between 0.5 and
1.5 per cent of the consumer gains from lower footwear prices.
6
Several studies are reviewed in WTO (2003).
There is significantly less evidence on the size of adjustment costs

in developing countries, but there are many reasons to expect that
the costs identified in the studies above would be much larger in
poorer countries. First, most of the industrial base of developing
economies is concentrated in a few key industries. In ‘one-industry
towns’ the costs of adjustment may be larger if the laid-off workers
from the primary industry cannot be absorbed into alternative
employment.7 For example, liberalization of the cashew market in
Mozambique in the late 1990s led to the loss of 85 per cent of the
workforce employed in the local processed-cashew industry. Recent
evidence suggests that whole towns have shut down as a result of the
factory closures (WTO 2002).
Adjustment costs may also be exacerbated in poor areas because of
limited access to credit. If capital markets are weak, viable firms
may not be able to finance the short-term costs associated with new
trade regimes, and laid-off workers may not be able to find funds to
retrain themselves for alternative jobs.
In many developing countries unemployment rates are high, and
accordingly, the length of time that individuals will spend unem-
ployed will be larger.8 Studies in the United States show that the
costs of adjustment for dislocated workers is lower for more-
educated workers—evidently their higher education makes them
more adaptable, and hence more mobile.9 With education levels
typically low in developing countries, one might expect the transi-
tion costs to be correspondingly greater.
There are other reasons why developing countries might suffer
larger adjustment costs than developed countries. This section
reviews some of the issues associated with the proposals emerging
from the Doha Round. In particular, the reduction of MFN tariff
rates will lead to the erosion of the preference margins currently
benefiting the exports of the least developed countries under various
non-reciprocal market access preference schemes.
Also, the reduction of tariffs has serious fiscal consequences for
many developing countries. Over thirty countries—mostly small
7
Among the costs that should be calculated are those associated with relocating labor, the loss in
the value of the housing capital stock, and the value of the associated infrastructure.
8
See Shapiro and Stiglitz (1984) for a calculation of the relationship between the level of unemploy-
9
ment and the cost of losing one’s job. For a survey of these studies, see Kletzer (2001).
and poor—derive more than 25 per cent of their public budgets from
tariff revenue. For these countries, trade liberalization will necessit-
ate massive reform of the taxation system to avoid fiscal crises.
Finally, developing countries face disproportionately high imple-
mentation costs from the proposals related to the Singapore Issues.
Regulatory agreements in areas such as trade facilitation and com-
petition policy require public expenditure on new laws, systems,
administration, and enforcement. Estimates of the costs of imple-
menting the regulatory changes in the Uruguay Round are high. For
developing countries, whose institutions are weakest and in greatest
need of reform to meet international standards, these implementa-
tion costs are disproportionately high. Thus, seemingly symmetric
rules may have asymmetric costs.
For these reasons, Doha Round agreements should offer equitable
market access rules as well as addressing differences in adjustment
costs in order to achieve a fair deal for developing countries.
Erosion of LDC trade preferences

Several developed countries offer non-reciprocal preferential market
access which reduces the tariff rates on the goods of least developing
countries below MFN rates. Almost 12 per cent of US imports
subject to MFN tariffs enter the US from LDCs under lower tariff
rates through such non-reciprocal preference programs. Many LDCs
fear that reductions in MFN tariff rates through multilateral trade
liberalization would harm their exports by eroding their preferential
margins.
Preferential tariffs for LDCs have formed an important part of
global trade architecture since the inception of the Generalized
System of Preferences (GSP) in 1968. Recently there have been a
number of initiatives in OECD countries to discriminate further in
favor of LDCs. Most notable among these are the EU’s Everything
but Arms (EBA) initiative and the US’s African Growth and
Opportunity Act (AGOA).
The net effect on LDCs of preference erosion through reduction
in MFN tariffs depends on whether the loss of trade diversion
(the negative switching or substitution that occurs as the margin

of their preferences declines) exceeds the gains from trade creation
(the increase in global trade resulting from improved market access).
The evidence below suggests that favorable trade diversion resulting
from preferences has had only a limited effect on most LDCs.
However, large effects on a small group of countries and a small
group of sectors cannot be ignored. The net effects of reductions in
MFN tariffs could be summarized as being positive and significant
for most industries in most countries (particularly developing coun-
tries outside the preferential schemes for least developed countries)
but large and negative for a small number of producers. The policy
implication of these results is straightforward: preference erosion is
not a consideration that should impede multilateral liberalization,
but it does suggest that compensation and adjustment programs
for the small group of net losers should be an integral part of any
liberalization program.
The benefits of trade preferences for LDCs
Preference schemes have been adopted in an effort to support the

development of poor countries and assist them to integrate into the
global trading system. Preferences increase the exports of benefi-
ciaries, partly by diverting trade from countries that do not receive
preferences. This competitive advantage may help LDCs to develop
through increased investment, employment, and growth. Additionally
preferences may encourage industrial diversification in countries
that have relied on the production of primary goods.
However, analysis of preferential schemes on LDC exports shows
only limited impact. Brenton (2003) studies the impact of the EBA
initiative, which the EU has argued will ‘significantly enhance
export opportunities and hence potential income and growth’ for
LDCs (CEC 2002). In 2001 the EBA initiative granted duty-free
access to imports of all products from the least developed countries
(except arms and munitions).10 Total exports from these LDCs to the
EU increased by 9.6 per cent in 2001. However, in practice, as noted
10
However, not all the preferences were implemented immediately; some will be delayed until 2009.
The calculations here ignore these future impacts.
earlier, the EBA was only relevant to the 919 products (of the EU’s
10,200 tariff lines) which had not previously been granted duty-free
status under either the GSP or the Cotonou Agreement.11 Of these
919 products, imports from LDCs were recorded in just 80 products
in 2001. Brenton (2003) notes that total exports of these products
actually fell from 3.5 million in 2000 to 2.9 million in 2001.12
Moreover, trade in these goods in 2001 amounted to just two-
hundredths of one per cent of the total value of LDC exports to the
EU. Thus it appears that the direct impact of the EBA initiative has
not been significant in the short term and, given the small size of
trade in affected products, it is not likely to be large in the medium
term. (Supporters of the EBA initiative are more optimistic; they
focus on the fact that its provisions are being implemented only
gradually over time. But as the discussion above indicates, the devil
is often in the detail, and this provides some grounds for skepticism.)
Analysis of the long-run effects of trade preference systems
including many countries requires general equilibrium analysis.
Laird, Safadi, and Turrini (2002) evaluate the effects of the GSP
scheme by analysing the welfare consequences of replacing GSP
with MFN tariff rates. Their computable general equilibrium
(CGE) simulations identify the costs and benefits of the GSP to a
range of countries. Table 13.1 reports the percentage change in
exports associated with the removal of the GSP (i.e. negative values
indicate benefits resulting from the GSP). The trade effects are
quite small, with the largest effect in South Asia, which is estim-
ated to have had an increase in exports of 1.58 per cent as a result of
preferences. There are also negligible effects on donor countries,
among whom Europe alone suffers a very small export decline.13
Table 13.1 also indicates that trade effects are concentrated in a
small number of sectors, particularly textiles and processed agri-
culture. The smaller effects in agriculture for some of the LDCs
11
Forty-four of these tariff lines were products such as bananas, rice, and sugar, for which liberalization
was delayed for up to 8 years.
12
The actual decline was even larger, but Brenton removes the effect of the large drop in EU imports
of Sudanese grain sorghum in 2000–1. These numbers do not include trade in the 44 products for which
liberalization was delayed. There may, of course, be other factors affecting trade, and it is clearly possible
that the declines would have been even larger but for the EBA initiative. The analysis above simply demon-
strates the limited scope of the agreement.
13
Note that this study only considers the GSP scheme and does not include the effects of other schemes
such as the EU’s Cotonou Agreement, which offers the largest gains to African countries.
Table 13.1. Export changes resulting from the replacement of GSP with MFN tariffs (GCE results, %)
Sectors
Exporting Mining Transportation Machines Metal Other Agricultural Agricultural Textiles, Services Total
region equipment manufacturing primary processed apparel
Asian NICs ⫺0.05 ⫺0.89 ⫺1.03 ⫺0.53 ⫺1.2 0.33 ⫺1.57 ⫺2.07 1.02 ⫺0.70
China ⫺0.13 1.09 ⫺0.31 ⫺0.53 ⫺1.0 0.21 ⫺1.45 ⫺2.67 0.91 ⫺0.8
South Asia 1.01 ⫺1.29 ⫺1.46 ⫺2.12 ⫺1.98 ⫺0.79 ⫺2.44 ⫺3.50 2.28 ⫺1.58
Western Europe ⫺0.60 ⫺0.20 0.13 ⫺0.36 0.23 0.01 0.62 1.40 ⫺0.67 0.001
North America ⫺0.70 0.03 0.06 ⫺0.11 0.05 ⫺0.62 0.06 0.14 ⫺0.49 ⫺0.12
CEED ⫺0.15 ⫺3.37 ⫺1.51 0.07 ⫺1.31 0.25 ⫺1.54 0.68 0.70 ⫺0.54
Sub-Saharan Africa ⫺0.06 1.78 ⫺1.00 2.27 ⫺1.70 ⫺1.43 ⫺9.40 ⫺5.96 1.77 ⫺0.75
Oceania ⫺0.68 0.50 0.17 ⫺0.31 0.01 ⫺0.49 ⫺0.02 1.13 ⫺0.09 ⫺0.16
North Africa/Middle East ⫺0.10 ⫺0.71 ⫺3.74 ⫺0.62 ⫺1.51 ⫺0.34 ⫺2.18 ⫺2.99 1.26 ⫺0.48
South America ⫺0.13 1.11 0.62 ⫺0.19 ⫺1.83 ⫺0.56 ⫺2.51 ⫺0.44 0.49 ⫺0.39
Japan ⫺1.47 ⫺0.67 ⫺0.40 ⫺1.13 ⫺0.79 ⫺1.37 ⫺1.33 ⫺4.34 ⫺0.63 ⫺0.64
Rest of world 0.145 ⫺1.83 ⫺2.88 ⫺0.24 ⫺2.80 ⫺0.48 ⫺6.90 ⫺7.02 1.58 ⫺1.46
Source: Laird, Safadi, and Turrini (2002).

Table 13.2. Welfare effects from the replacement of GSP with MFN tariffs (GCE results, US $m)
Allocation Terms-of-Trade Total % change

component component
Asian NICs ⫺405 ⫺1,950 ⫺2,317 ⫺0.23

China ⫺360 ⫺1,613 ⫺1,855 ⫺0.15
South Asia ⫺327 ⫺594 ⫺964 ⫺0.19
Western Europe ⫺722 4,634 3,719 0.05
North America 85 1,866 2,252 0.02
Transition economies ⫺317 ⫺941 ⫺1,297 ⫺0.17
Sub-Saharan Africa ⫺173 ⫺512 ⫺701 ⫺0.22
Oceania 1 ⫺22 ⫺11 ⫺0.003
North Africa and Middle East ⫺474 ⫺1,315 ⫺1,816 ⫺0.23
Latin America ⫺226 ⫺789 ⫺1,043 ⫺0.05
Japan ⫺246 1,466 1,189 0.033
Rest of world ⫺107 ⫺256 ⫺446 ⫺0.17
Total ⫺3,275 ⫺27 ⫺3,293
Source: Laird, Safadi, and Turrini (2002: table 5).
suggest that the GSP might play an important role in diversifying

the industrial base of those economies. Table 13.2 reports the
effects of the GSP on welfare. The largest beneficiaries in percent-
age terms are Africa and the Asian NICs. Overall the effects are
quite small, amounting to approximately 0.2 per cent of real income
in any region. The implication of these results, if correct, is that
there is likely on average to be little difference between the impact
of trade liberalization measures (which undermine the benefits of
preferences) on countries that are the beneficiaries of preferences
and those that are not.
Why are the benefits so small?

In practice LDCs are often not able to realize much of the benefit
promised by market access preferences. This is evident in the low
degree of utilization of preference schemes. Table 13.3 illustrates the
utilization of preferences offered by Canada, the EU, Japan, and the
US. The table separates the underutilization of preferences into a
component relating to the generosity of the scheme itself (the
Table 13.3. Utilization of non-reciprocal preferences granted by the Quad countries to LDCs, 2001
Imports Imports Product

Total Dutiable eligible receiving coverage Utilization Utility
imports imports for GSP GSP percentage percentage percentage
(1) (2) (3) (4) ((4)/(3)) ((4)/(2))
Canada 243.2 94.6 11.4 8.0 12.1 70.2 8.5

EU 4372.4 3958.1 3935.7 1847.4 99.4 46.9 46.7
Japan 1001.3 398.1 278.3 228.4 69.9 82.1 57.4
US 7221.3 6716.3 2960.1 2836.1 44.1 95.8 42.2
Total 12838 11167.1 7185.5 4919.9 64.3 68.5 44.1
Source: UNCTAD (2003).
product coverage ratio) and a component relating to the take-up rate

(utilization ratio).
Interestingly, Table 13.3 shows that the EU has a high product
coverage percentage and a lower than average utilization percentage
and the US seems to have the reverse. In the case of the EU, over 50
per cent of eligible exports are not getting preferential access. Part of
the reason for this is stringent rules of origin which are designed to
prevent trade deflection, whereby products from non-beneficiary
countries are routed through LDCs to exploit the preferences.
Brenton (2003) suggests that one reason for the lack of take-up is that
it can often be difficult or costly to acquire the required documenta-
tion to satisfy rules of origin.
Recent literature suggests that rules of origin are a main reason
for the under-utilization of trade preferences (see Estevadeordal
2000), and preference-receiving countries themselves consistently
identify rules of origin as a problem for their exporters.14 Rules of
origin often require exporters to devise and operate a new account-
ing system, which in most cases differs from existing systems
designed to deal with domestic legal requirements. In many cases the
additional expenditure incurred in operating a parallel accounting
14
The shortcomings of the origin system and consequent obstacles to the utilization of preferences
identified by preference-receiving countries were discussed in the context of the UNCTAD Working Group
on Rules of Origin and in the Special Committee on Preferences. See ‘UNCTAD, Compendium of the work
and analysis conducted by UNCTAD Working Groups and Sessional Committees on GSP Rules of Origin,
Part I’ (UNCTAD/ITD/GSP/31) of 21 February 1996.
system may outweigh the benefit conferred by tariff preferences.

This possibility is supported by evidence suggesting that under-
utilization is strongest in sectors where the preference margin is
lowest and therefore less likely to be greater than the administra-
tion costs of complying with rules of origin.15 Another cost imposed
by rules of origin is that they may disqualify LDC exporters from
preferences if they avail themselves of cheap imported intermedi-
ate inputs. UNCTAD (1996) finds evidence of this by relating peaks
in Bangladeshi and Cambodian imports of fabrics from China to
low utilization rates of preference schemes by textile exporters in
those countries. This can be assumed to be a strong indication that
the manufacturers have chosen to give up tariff preferences
because they cannot comply with rules-of-origin requirements.
Another problem with preference schemes is that they are not
particularly generous when rates are weighted across goods.
Table 13.4 shows the benefits of GSP rates against MFN rates. When
simple averages are taken the GSP rates are more favourable in all
countries. However when weighted across goods average MFN rates
are even lower than average GSP rates in the EU, Canada, and Japan.
This occurs because the GSP rates in those countries are set as a
margin under the MFN rates which are typically higher on imports
Table 13.4. Tariff averages for imports under MFN and GSP, 1999
Simple tariff average Weighted tariff average
Canada MFN 4.49 1.27

GSP 2.89 4.18
Japan MFN 5.28 1.97
GSP 2.2 3.47
US MFN 5.59 2.56
GSP 0.0 0.0
EU MFN 7.07 3.56
GSP 5.23 4.54
Source: Laird, Safadi, and Turrini (2002)
15
UNCTAD (1996): 58.7 per cent of those Mexican exports to the US that were eligible for preferences
but were not imported under them constituted goods whose preference margin was less than 5 per cent.
Evidently, the cost of establishing that one is qualified to receive preferential treatment exceeds the benefits.
from developing countries. That means that MFN rates are so much
higher on the goods exported by developing countries that even after
GSP discounts, LDCs face higher average tariff rates. In a sense,
the GSP only partially compensates for the discrimination by
developed countries against the goods produced by the developing
countries.
Figures 13.1a and 13.1b show the import duties paid to the US on
high-value agricultural products. Figure 13.1a indicates that the US
tariff system already discriminates against both low-income and
middle-income countries in these products. Whereas high-income
countries account for 75 per cent of dutiable imports in 2001 they
paid only 56 per cent of duty. By contrast middle- and low-income
countries paid higher duty as a proportion of dutiable imports. The
reason is that (even allowing for preferences) US tariff peaks tend to
concentrate on goods exported intensively by developing countries.
Figure 13.1b calculates an ‘effective tariff rate’ as the ratio of duties
collected to the value of imports. The effective tariff rate paid by
middle-income countries is almost twice as much as that paid by
developed countries. Trade preferences often further discriminate
against middle income countries.
(a) 80 (b) 10
Share of dutiable imports
Share of import duties paid 8
60
6
40
4
20
2
0 0
High-income Middle-income Low-income High-income Middle-income Low-income
countries countries countries countries countries countries
Figure 13.1. Agricultural import protection in the US

(a) Share of dutiable high-value agricultural products imports to the US and
share of duty paid, 2001 (% of total)
(b) Effective US import tarrifs on high-value agricultural products by source,
2001 (% ad valorem)
Source: Wainio and Gibson (2003).
The impact of the erosion of the benefit of trade preferences

as a result of lower MFN rates
Estimates of the benefits of preferences for LDCs (often calculated

as the costs LDCs would experience if they were eliminated) are
different from estimates of the costs of preference erosion through
reduced MFN tariff rates. The chief difference is that in the case of
preference erosion LDCs are compensated for the loss of competitive
advantage in donor countries by increased market access in all other
countries. As a result the costs to LDCs of preference erosion
through MFN tariff reductions are likely to be smaller than the costs
of preference elimination.
Waino and Gibson (2003) use a partial equilibrium model to
estimate the effects of tariff-cutting liberalization on developing
country exports. Their study focuses on the US and considers only
a range of high-value agricultural products, including fresh fruits
and beverages and processed food and beverages. Table 13.5 reports
the results of an experiment in which all tariffs are eliminated. In
particular it reports the trade creation effects (derived from the
price reductions following tariff cuts) and the trade-shifting effects
(from domestic consumers switching to those goods whose
protection—and price—has declined most) for countries receiving
non-reciprocal preferences, free-trade partners, and countries sub-
ject to MFN tariff rates.16 The table indicates that the magnitude of
trade effects is small. It also indicates that the beneficiaries of
non-reciprocal trade preferences suffer no net loss from loss of prefer-
ences. Their market share is reduced but this is offset by capturing
additional market share in those commodities for which they face
(the now-lower) MFN tariffs, but also compete with exporters in free
trade agreements.
Waino and Gibson’s study probably understates the net benefits of
liberalization because they do not include the gains to LDCs from
increased market access in countries where they were not previously
receiving benefits. Since 28 per cent of LDC exports go to developing
countries, and developing countries have higher average MFN tariffs
16
Quite simply the model assumes that tariff reductions flow through to price reductions and increase
the level of US domestic demand. However, the trade creation effects in their model may be overstated
since they assume an infinite elasticity of supply.
Table 13.5. Effect of full tariff liberalization on high-value agricultural imports to US (US $m)
Preference FTA MFN Global

beneficiaries partners partners total
Pre-liberalization imports 4,503 9,709 7,551 21,762

Effect of liberalization
Trade created 194 92 391 679
Trade shifted ⫺23 ⫺153 176 0
Total effect 171 ⫺61 567 679
Total effect as percentage 3.8% ⫺0.6% 7.5% 3.1%
Source: Waino and Gibson (2003).
than developed countries, the lowering of these tariffs is a poten-

tially significant source of trade creation.
Waino and Gibson (2003) offer the example of Peruvian asparagus,
US$26.7 million of which was exported to the US at preferential
rates and US$5.9 million at MFN rates. Peru is a low-cost exporter
of asparagus and thus the reduction in its margin of preference
caused only a small loss of market share, which would be more than
offset by increased MFN exports under lower rates.17 Moreover, if
liberalization includes reductions in quotas, countries receiving
preferences may even experience an increase in their exports to
donor countries.
Vulnerable industries
The results above indicate that the average effect of preference

erosion on LDCs is unlikely to be large. However, this is not true for
all industries in all countries. Industries that are particularly reliant
on preferences could be seriously damaged by preference erosion.
In general the higher the dependency of countries on preferences,
the larger the potential loss from MFN tariff cuts. Table 13.6 gives a
useful insight into developing countries’ levels of EU preference
dependence. The penultimate column gives the share of exports
newly liberalized under the EBA in 2001 as a percentage of total
17
In a situation such as that just depicted, the marginal export is at MFN rates; the impact on trade is thus
determined solely by the change in the MFN rate. Peruvian exporters, however, lose a rent equal in value to
the value of the preference.
Table 13.6. Importance of Products Liberalized under the EBA (monetary values in US$ 000)
Total exports EBA exports Sugar, EBA Sugar, bananas,

to EU (products liberalized bananas, exports share and rice export
in 2001) and rice (%) share (%)
(1) (2) (3) ((2)/(1)) ((3)/(1))
ACP countries
Angola 1,944,630 91 0 0.00 0.00
Congo 941,784 7 50 0.00 0.01
Equatorial Guinea 754,865 0 0 0.00 0.00
Liberia 736,973 10 0 0.00 0.00
Madagascar 600,912 72 8,500 0.01 1.41
Guinea 579,518 41 0 0.01 0.00
Mozambique 530,174 248 991 0.05 0.19
Tanzania 395,283 35 6,648 0.01 1.68
Sudan 303,550 778 13,982 0.26 4.61
Mauritania 258,568 6 6 0.00 0.00
Uganda 242,524 116 55 0.05 0.02
Malawi 194,903 0 22,617 0.00 11.60
Ethiopia 159,389 12 968 0.01 0.61
Zambia 158,375 1,359 6,675 0.86 4.21
CAR 152,804 0 0 0.00 0.00
Niger 119,613 6 0 0.00 0.00
Benin 63,698 69 0 0.11 0.00
Burkino Faso 63,052 52 0 0.08 0.00
Djibouti 61,494 38 0 0.06 0.00
Togo 58,591 26 26 0.04 0.02
Chad 57,638 1 0 0.00 0.00
Mali 45,726 67 0 0.13 0.00
Sierra Leone 38,420 72 0 0.19 0.00
Rwanda 21,782 6 78 0.03 0.36
Comoros 20,770 3 0 0.00 0.00
Gambia 20,679 41 0 0.00 0.00
Burundi 19,474 19 0 0.10 0.00
Lesotho 12,797 0 0 0.00 0.00
Haiti 16,356 158 0 0.97 0.00
Vanuatu 13,653 0 0 0.00 0.00
Cape Verde 11,803 10 0 0.11 0.00
Sao Tome 8,009 0 0 0.00 0.00
Eritrea 6,737 1 0 0.01 0.00
Solomon Islands 4,975 0 0 0.00 0.00
Guinea Bissau 4,542 0 0 0.00 0.00
Somalia 3,047 0 0 0.00 0.00
Samoa 2,206 0 0 0.00 0.00
Kiribati 728 0 0 0.00 0.00
Table 13.6. continued
Total exports EBA exports Sugar, EBA Sugar, bananas,

to EU (products liberalized bananas, exports share and rice export
in 2001) and rice (%) share (%)
(1) (2) (3) ((2)/(1)) ((3)/(1))
Tuvalu 390 0 0 0.00 0.00

Non-ACP countries
Bangladesh 3,318,865 69 5 0.00 0.00
Cambodia 482,480 0 0 0.00 0.00
Laos 143,716 74 42 0.05 0.03
Nepal 135,119 0 0 0.00 0.00
Yemen 83,596 169 0 0.20 0.00
Maldives 37,377 1 0 0.00 0.00
Afghanistan 23,813 0 0 0.00 0.00
Bhutan 552 0 27 0.00 4.89
Total 12,859,883 3,658 60,670 0.03 0.47
Total-ACP 8,634,365 3,344 60,596 0.04 0.70
Total Non-ACP 4,225,518 313 74 0.01 0.00
Source: Brenton (2003).
exports to the EU for a range of developing countries. The average is

0.03 per cent, and no country is higher than one per cent. This indicates
that the amount of products involving preferences is a small fraction
of total exports and that the erosion of preferences is unlikely to have
a significant affect on these countries.
The table also reports the proportion of exports in a group of
vulnerable industries, including bananas, rice, and sugar (final
column). The share of these goods in total exports is quite large
for a small group of countries, reaching a peak of 11.6 per cent in
Malawi. Unfortunately, tariffs on these goods were not elimin-
ated immediately under the EBA in 2001, but they will gradually
be phased out by 2009. In these critical goods, the increased pref-
erence margin potentially has a large effect on exports and, con-
versely, the erosion of these preferences by MFN tariff reduction
could have a large negative effect on these countries.
Addressing the problems of adjustment in critical industries
in vulnerable countries should be a key component of any
multilateral reform proposal. There are many examples of critical
industries—particularly in small countries—which face highly

negative consequences from preference erosion.
Assistance for critical industries and their workers is a preferred
solution to the maintenance of preference margins. There are two
reasons to prefer assistance to delayed MFN liberalization. First,
delayed liberalization discriminates against developing countries
which do not benefit from preferences. The second reason for prefer-
ring assistance is that the maintenance of long-term preferences
induces beneficiaries to specialize in activities in which they may
never be competitive once preferences are removed. This discourages
industrial diversification and increases adjustment costs when the
preferences are eventually removed.
At the same time, it should be recognized that sometimes, provid-
ing even temporary preferential access can provide long-term gains.
By excluding some critical products (particularly bananas, rice, and
sugar) from immediate zero tariff under the EBA in 2001, the EU may
be missing the opportunity to provide these industries with a
foothold in their markets in advance of MFN liberalization.18
Fiscal effects
In some countries tariff revenues make up a substantial part of total
government revenue. Many of these countries are concerned that
trade liberalization will have a significant adverse effect on public
revenue and the ability to fund public expenditure.
Taxes on international trade account for around one per cent of
government revenues in developed countries and around 30 per cent
in the least developed countries. Small countries are the most reliant
on tariffs. For example, tariffs make up 62 per cent of tax revenue in
the Bahamas, 54 per cent in the Solomon Islands, and 75 per cent in
Guinea (Ebrill, Stotsky, and Gropp 1999). Figure 13.2 shows the ratio
of tariff revenue to GDP for five country groups. African govern-
ments are most reliant on revenue from tariffs, followed by Middle
18
Though, to be sure, for some of these products, it is unlikely that there will be MFN liberalization any
time soon. Moreover, market ‘loyalty’ is likely to be less important in ‘commodity’ trade than in trade in
manufactures.
4
% of GDP
0
OECD South America Asia and Pacific Middle East Africa
Figure 13.2. Average tariff revenue, 1995 (% of GDP)

Source: Ebrill, Stotsky, and Gropp (1999).
Eastern and Asia/Pacific countries. Table 13.7 shows tariff revenue

as a proportion of GDP for 8 countries. The table demonstrates that
changes in tariff revenue resulting from trade reform have wildly dis-
proportionate effects on developing countries.
The complete elimination of tariffs would, of course, reduce tariff
revenue to zero. This scenario would require developing countries to
change fundamentally the structure of their taxation systems in
order to raise revenues from other sources.
However, in practice the effect of less ambitious trade liberaliza-
tion on government revenues is more complex. If government
replaces tariffs with, for instance, a uniform VAT imposed at the
point of production—and for imported goods, at the point of
importation—then the only difference is that tariffs would be taxed
at a uniform rate (and domestic production would be taxed at the
same rate). Typically, governments also provide an export rebate, so
that the import content of exported goods is effectively tax-exempt.
Moreover, trade liberalization often involves a range of reforms
other than tariff reductions, such as the elimination of trade-related
subsidies and the ‘tariffication’ of non-tariff barriers. Many of these
reforms could increase government revenues (see Table 13.8).
For these reasons the effect of trade liberalization on government
revenues is difficult to predict. Senegal pursued trade liberalization in
the mid-1980s, following which there were large revenue shortfalls.
Lost tariff revenue combined with slow growth in trade volumes and
Table 13.7. Tariff revenue for selected countries, 1995 (% of GDP)
OECD Asia/Pacific South/Central Middle East Africa

America
Average 0.37 Average 3.42 Average 2.36 Average 3.48 Average 5.39
Australia 0.65 Fiji 4.95 Argentina 0.70 Bahrain 2.79 Botswana 5.85
Austria 0.14 India 2.76 Bahamas 8.32 Egypt 3.59 Burundi 2.51
Belgium 0.52 Indonesia 1.03 Bolivia 1.05 Iran 2.34 Cameroon 2.56
Canada 0.42 Korea 1.32 Brazil 0.46 Israel 0.19 Congo, Democratic 1.33
Republic of
Denmark 0.20 Malaysia 3.04 Chile 2.20 Jordan 7.74 Côte D’Ivoire 6.17
Finland 0.23 Myanmar 0.89 Colombia 1.43 Kuwait 0.95 Ethopia 2.30
France 0.13 Nepal 3.00 Costa Rica 2.97 Morocco 4.27 Gabon 4.47
Germany 0.21 Papua New 3.85 Dominican 3.80 Oman 0.95 Gambia 8.76
Guinea Republic
Greece 0.23 Philippines 4.96 Ecuador 1.64 Pakistan 4.88 Ghana 3.37
Iceland 0.38 Singapore 0.35 El Salvador 2.12 Syria 2.48 Kenya 3.94
Ireland 0.51 Solomon 11.43 Guatemala 1.87 Tunisia 8.14 Lesotho 32.27
Islands
Italy 0.15 Sri Lanka 3.68 Nicaragua 5.25 Malawi 3.02
Japan 0.21 Thailand 3.15 Panama 2.62 Mauritius 6.74
Mexico 0.61 Paraguay 1.75 Rwanda 3.22
Netherlands 0.59 Peru 1.54 Senegal 4.81
New Zealand 0.92 Uruguay 0.96 Sierra Leone 3.62
Norway 0.30 Venezuela 1.39 South Africa 0.18
Portugal 0.28 Zambia 3.06
Spain 0.19 Zimbabwe 4.30
Sweden 0.27
Switzerland 0.33
Turkey 0.76
UK 0.35
US 0.27
Source: Ebrill, Stotsky, and Gropp (1999).
weaknesses in economic management led to dire fiscal consequences.

To raise more revenue, the tariff reductions were quickly abandoned
and the liberalization process delayed. By contrast, trade liberaliza-
tion in Morocco was accompanied by programs to broaden the
domestic tax base, including the introduction of a VAT in 1986. As a
consequence, Morocco was able to reduce its reliance on trade taxes
while maintaining a stable ratio of public revenue to GDP.
The desirability of replacing revenue from trade taxes with domestic
revenue sources raises the issue of relative efficiency of alternative
Table 13.8. Summary of effect of trade liberalization on revenue
Trade reform Expected revenue effect
Replace NTBs with tariffs Positive

Eliminate tariff exemptions Positive
Eliminate trade-related subsidies Positive
Reduce tariff dispersion Ambiguous/Positive
Eliminate state trading monopolies Ambiguous/Positive
Reduce high average tariffs Ambiguous
Lower maximum tariff Ambiguous
Eliminate export taxes Ambiguous/Negative
Reduce moderate or low average tariffs Negative
Source: Sharer et al. (1998).
forms of taxation. There is some theoretical research suggesting that

reducing trade taxes and replacing them with a consumption tax is
welfare-enhancing (Keen and Lightart 1999) on the basis that they
are broader and less distortionary. More recently, however, Emran
and Stiglitz (2004) have shown that in developing countries with an
informal sector in which, say, a VAT cannot be imposed, it is desir-
able to retain some trade taxes, e.g. to tax imports at a higher rate
than domestic production.
The issues of complementary policies to minimize the fiscal
effects of trade reforms will be taken up later in the chapter.
The main point in this section is that global trade reform has
significant consequences for the fiscal structures of developing
countries, whereas developed countries are by and large immune.
Developing countries are likely to suffer either a loss of total tax
revenue or, at best, a large administrative cost—and even more
economic distortions—associated with the implementation of a
new taxation system.
Implementation costs
While traditional market access agreements such as tariff and quota
reductions incur small implementation costs, the ‘new’ trade
agenda embodied in the Uruguay Round and even more in the
Singapore Issues may impose a much larger implementation burden.

Implementation costs are another example of how WTO agreements
may have different impacts on poor and rich countries. Compliance
with WTO agreements is harder for developing countries, whose
administrative systems usually require larger reform to meet agreed
standards. In addition, developing countries have the weakest
government institutions and the greatest constraints on public
resources. Implementation of an agreement incorporating the
Singapore Issues would require expenditure on system design and
drafting of legislation; capital expenditure on buildings and equip-
ment; personnel training; and the ongoing costs of administration
and enforcement.
Finger (2000) points out that the implementation of regulatory
agreements will often draw money from the development budgets of
poor countries. For this reason such agreements should be analysed
in terms of their rate of return and compared to the alternative devel-
opment priorities on which the same money could be spent. Finger
estimated the implementation of three of the Uruguay Round’s six
agreements that required regulatory change (customs reform, intel-
lectual property rights, and sanitary and phytosanitary (SPS) mea-
sures). His analysis suggests that the average cost of restructuring
domestic regulations in the twelve developing countries considered
could be as much as US$150 million. In eight of these countries this
figure is larger than the entire annual development budget.
Many developing countries have been unable to meet their
Uruguay Round obligations because of these high costs. By January
2000, up to 90 of the WTO’s 109 developing country members were
in violation of the SPS, customs valuation, and TRIPS agreements.
Estimates of the cost of compliance with the Uruguay agreements
vary widely depending on the quality of the existing systems and the
strength of institutions in each country. Hungary spent more than
US$40 million to upgrade the level of sanitation of its slaughterhouses
alone. Mexico spent more than US$30 million to upgrade intellectual
property laws. Finger (2000) suggests that for many of the least devel-
oped countries in the WTO compliance with these agreements is a
less attractive investment than expenditure on basic development
goals such as education.
The costs of implementing the regulatory agreements that could

potentially emerge from the Doha Round will vary widely across
countries. However, many of the proposed reforms within the
Singapore Issues could be costly. For example, were there to be new
competition regimes (which seems unlikely), such regimes would be
difficult to implement. Competition law is technical and requires
institutional skills and resources that are in short supply in many
developing countries. In addition, competition law enforcement is
expensive. OECD and national sources indicate that the annual
budget of the antitrust office in OECD countries is in the US$15–50-
million-plus range. For developing countries with enforcement
agencies the budgets are lower but still significant (Hoekman and
Mavroidis 2002).19
Similarly the costs of trade facilitation could be large for some
countries. For example, the World Bank assisted Tunisia in its
program of streamlining and modernizing its customs procedures.
The total value of World Bank loans to Tunisia for this purpose was
US$35 million in 1999. Similarly the World Bank lent US$38 million
to Poland for upgrading the physical and managerial infrastructure of
its port facilities (Wilson 2001). Projects to implement the WTO
Agreement on Customs Valuation, which also includes broader
customs reform, have been estimated to cost between US$1.6 million
and US$16.2 million. For example, a six-year program in Tunisia to
computerize and simplify procedures cost an estimated US$16.2
million (Finger and Schuler 2000). However, Bolivia implemented a
broad customs reform programme that cost US$38.5 million.
The size of the implementation costs associated with the
Singapore Issues raises questions about the appropriateness of their
inclusion in the Doha agenda. The important lesson from the
Uruguay Round is that regulatory changes imposed a large and (in
the case of the many non-compliant countries) unacceptable burden
on developing countries. The rules seemed to be constructed with
little awareness of development problems and little appreciation for
the institutional capacities of least developed countries.
19
Note, for example, the costs of antitrust offices in Mexico (US$14 m), Poland (US$4.1 m), Hungary
(US$2 m), and Argentina (US$1.4 m). There are doubts about whether these sums provide adequate
enforcement.
Poverty and labor markets

It is not an easy task to identify social groups who systematically suf-
fer as a result of trade liberalization. Heterogeneous trade patterns
and factor endowments across countries mean that similar reform
scenarios would have different effects on similar social groups in dif-
ferent countries, and hence the consequences of reform are often
masked in cross-country data.
Predicting which social groups in which countries will be detri-
mentally affected by trade reform is a task that defies generalizations
about the consequences for ‘the poor’. Instead, identifying the losers
from trade reform requires analysis of the particular effects on different
groups stratified by income source and expenditure patterns.
There are a variety of effects on workers—both wages and unem-
ployment rates may be affected; and the effects in the short run may
differ markedly from the effects in the long run. In the short run,
there are a number of reasons that markets do not adjust quickly, so
that the impact of liberalization is that workers lose jobs; this job
loss occurs faster than job creation. This is especially the case in
developing countries where financial markets are weak (so that
firms cannot quickly take advantage of any new opportunities that a
new trade agreement opens up), or when the country has a tight
monetary policy (e.g. as part of a so-called structural adjustment pro-
gram). But even if job creation matches job destruction, the new jobs
may require different skills, and may be created in different locales.
Typically, dislocated workers, even in advanced industrial countries
with low unemployment rates, suffer marked reductions in their
wages (part of which, but only a part, may be explained by a loss of
rents from being in a protected sector).
Economic theory suggests that if the economy manages to remain
at full employment then trade liberalization will tend to lead to factor
market price equalization, i.e. higher wages for unskilled labor in
less developed countries and lower wages in developed countries.
Incomes on average may increase, as countries are able to exploit
their comparative advantage. But even this conclusion has been
questioned, as attention has focused on the consequences of market
imperfections, such as imperfect competition or incomplete risk
markets. For instance, as Newbery and Stiglitz (1982) showed, trade

liberalization may make everyone worse off when risk markets are
limited.
Even when trade liberalization leads to increased efficiency, it is a
one-off effect. It does not necessarily lead to sustained increases in
the rate of growth of productivity, and indeed, if trade liberalization
is associated with greater volatility, there is the possibility that it
will actually lead to slower productivity growth.
A number of empirical studies have, nonetheless, tried to argue
that trade openness leads to increased productivity growth (e.g. Sachs
and Warner 1995; Sala-i-Martin 1997), and there is also evidence that
increased productivity growth leads, in the long run, to increased
wages (though there may be differential effects on the wages of
particular groups, even negative effects). However, both of these
relationships are controversial (Rodriguez and Rodrik 2000). Part of
the problem with the studies focusing on the relationship between
trade openness and productivity growth is that they are beset by a
host of econometric and interpretive problems. Openness itself, for
instance, is an endogenous variable, particularly as it is often
measured (e.g. as ratio of trade to GDP).
An example of the problems in interpretation is provided by a
recent study by Rama. Rama (2003) classifies seventy countries into
three groups: rich countries, non-globalizers (not yet fully integrated
into the international market), and recent globalizers.20 Figure 13.3
reports the growth rate of the average wage between the 1980s and
the 1990s for a set of common occupations. He suggests that this
implies that openness is good for workers, at least at an aggregate
level. But the countries that are not integrated into the international
market include those with a host of other problems—African coun-
tries, for instance, facing civil strife or the AIDS epidemic. These other
circumstances (inadequately controlled for in the statistical analysis)
may provide more of an explanation for the poor performance—in
trade, growth, productivity, and wages—than ‘trade openness’.
Francois, van Meijl, and van Tongeren (2003) use a general
equilibrium framework to analyse the effects of three alternative
20
Rama uses the three-group classification proposed by Dollar and Kray (2001).
35
30
25
% growth
20
15
10
5
0
Non-globalizers Rich countries Globalizers
Figure 13.3. Wage growth by country groups, 1980s–1990s

Source: Freeman, Oostendorp, and Rama (2001).
liberalization scenarios, representing a range of possible outcomes

from the Doha Round.21 The pattern of unskilled wage changes across
countries is illustrated in Figure 13.4. Wages increase for all countries
with the exception of China and some Central and Eastern European
countries. China is hurt because the authors build China’s full
accession to the WTO into their baseline. Further global liberaliza-
tion increases competition for exports and reduces world prices for
low-skill manufactures. The effect of liberalization increases over
time and the more ambitious the reform scenario, the larger the
benefits. But as we noted earlier, these general equilibrium models
need to be used with caution. They typically do not incorporate a host
of market imperfections that characterize developing countries, and
they almost never incorporate dynamic changes (e.g. those associated
with the adoption of new technologies). China may be in a better
position to grab and maintain market share than other developing
countries, in which case the results could be quite the opposite.
Average wage data may conceal more complex effects of trade
liberalization. Disaggregated analysis reveals that workers in some
sectors may gain while others lose. If inequality rises sufficiently,
21
Experiment 1 models a linear 50% reduction in all forms of protection, including agricultural and
industrial tariffs, export subsidies, OECD agricultural domestic support, and tariff-equivalent barriers.
Experiment 2 models a ‘Swiss formula’ reduction in which the maximum tariff is reduced to 25% (see
Francois and Martin 2003). Experiment 3 models a complete elimination of protection.
experiment 3, long run Rest of world

experiment 3, short run Sub-Saharan Africa
experiment 2, long run South Africa
experiment 2, short run
Australia/New Zealand
experiment 1, long run
experiment 1, short run Other Asia Pacific
High Income Asia
India
China
South America
North America
Mediterranean
CEECs
Rest of EU 15
Germany
France
Netherlands
–20.00 –10.00 0.00 10.00 20.00 30.00 40.00
Figure 13.4. Changes in unskilled wages resulting from three liberalization

alternatives within the Doha Round
Source: Francois, van Meijl, and van Tongeren (2003).
the poor’s gain from overall per capita income may be offset.
Certainly there are many well-known cases of countries where
inequality has risen as they have become more integrated into the
world economy. Inequality increased in Argentina, Chile, Colombia,
Costa Rica, and Uruguay after they liberalized trade at different
times (World Bank 2000). The decade following the signing of
NAFTA saw real wages in Mexico fall, even as trade increased.22
These results seem to contradict standard economic theory (the
Samuelson–Stolper theorem), which predicts that in countries with
a relative abundance of unskilled labor, trade liberalization should
result in a reduction in inequality. But this re-emphasizes the points
made earlier—the importance of market imperfections, including
the absence of risk markets, and dynamic effects. In some cases,
trade liberalization has exposed countries to more risk, and the poor
often bear the brunt of such risk. In other cases, greater global
22
Such data does not, of course, address the counterfactual: what would have happened but for the
trade agreement? But it certainly shows clearly that trade liberalization by itself is no guarantee of improved
living standards for workers.
integration leads to an increased transfer of technology, and if these

include unskilled-labor-saving innovations, they may well lead to a
lowering of unskilled real wages.
As Figure 13.5 indicates, while there does not appear to be a strong
relationship between income and wage inequality and openness,23
different types of liberalization shocks have different effects on dif-
ferent income groups. For example, there is considerable evidence
that financial liberalization can expose an economy to shocks which
are particularly pernicious for the poor. Levinsohn, Berry, and
Friedman (1999) examine how the 1997–8 Indonesian economic
crisis affected the poorest households. Through a cost-of-living
analysis they concluded that the lowest-income households tended
to be hurt the most. By contrast, Minot and Goletti (2000) analyse
the effects of another kind of liberalization shock on the well-being
of the poor. They examine how rice market liberalization in Vietnam
(principally the removal of quota restrictions) may affect poverty
(a) 1.4
(b)
3
Average annual change in Gini coefficient
1.2
Standard deviation of log wages
2
1
1
0.8
–0.06 –0.04 –0.02 0.02 0.04 0.06 0.08
0.6 –1
y = 2.522x + 0.0139
0.4 –2
R2 = 0.0013
0.2 –3
–4
0
0 50 100 150 200 250 300 Average annual change in trade/GDP
Foreign trade as % of GDP
Figure 13.5. Liberalization and inequality

(a) Wage dispersion and openness to trade
Source: Rama (2003).
(b) Change in trade and inequality
Source: Dollar and Kray (2001).
23
Rama (2003) argues that wage inequality across occupations does not increase with openness.
Others (e.g. Dollar and Kray 2001) have suggested that there is no discernible relationship at the country
level between trade openness (measured by trade volumes) and inequality (measured by the Gini coeffi-
cient). However, as noted earlier, openness (as measured) is an endogenous variable, so that the results
have limited value in providing inferences concerning the effects of a change in policy, such as liberalization.
levels. They find that liberalization raises prices and that since rice
production is relatively labor-intensive in Vietnam, a rise in prices
increases the demand for agricultural labor and consequently the
agricultural wage rate. They find that the net effect on real incomes
in rural areas is generally positive and reduces most measures of
poverty. These examples serve to indicate that different liberalization
programs may have different effects on poverty which are masked in
cross-country data.
These examples make clear that a full analysis of the impact on
poverty must look at impacts on unemployment, factor prices, and
goods prices. As previously noted, computable general equilibrium
(CGE) models are frequently used to analyse the effects of trade
liberalization on factor and consumer prices across countries.
Unfortunately, they seldom incorporate risk and unemployment,
which play an important role in generating poverty. In addition,
most of these models consider the welfare change for a single repres-
entative household in each region, making them a poor tool for the
analysis of poverty. Hertel, Ivanic, Preckel et al. (2003a) augment the
standard type of CGE analysis by adding several different types of
households categorized by their income source into households
whose primary source of income is from (1) transfers, (2) agriculture,
(3) non-agricultural business, (4) wages, or (5) diversified sources.
They use survey data for several countries to include information on
the income and expenditure profile of each group. This framework
allows them to look at the effects of trade liberalization at a sub-
national level to discover vulnerable populations within countries
whose plight might have been masked in national-level data.
They find that poverty rates do not fall uniformly within coun-
tries. Table 13.9 shows the model’s price change predictions for
Indonesia (Hertel, Ivanic, Preckel et al. 2003a). Increased demand for
Indonesia’s exports bids up their price relative to the world average.
The home price of commodities rises (see the short run total column)
as other countries reduce their protection and the EU and US reduce
the supply of subsidized exports (see the Liberalization by DCs col-
umn). The price rise is not offset by the cuts in the relatively modest
Indonesian agricultural tariff rates (Own-country liberalization
column). By contrast the prices of manufactured goods (durables and
Table 13.9. Effect of global trade liberalization on market prices in Indonesia (% change)
Short run Long run
Own-country liberalization Liberalization by DCs Liberalization by LDCs Total Total
Agricultural Non-agricultural Agricultural Non-agricultural Agricultural Non-agricultural
Factors
Agricultural Profit ⫺0.8 ⫺1.7 3.9 2.0 0.1 0.0 3.5 Land 5.3
Non-Agricultural Profit ⫺0.3 ⫺0.1 1.5 3.2 0.1 ⫺0.3 4.1 Capital 5.0
Unskilled labor ⫺0.4 0.0 2.1 3.4 0.1 0.1 5.3 Unskilled wages 6.1
Skilled labor ⫺0.4 0.1 1.6 3.3 0.1 0.1 4.8 Skilled wages 5.3
Public transfers ⫺0.5 ⫺0.2 2.0 3.1 0.1 ⫺0.2 4.4 Public transport 5.7
Private transfers ⫺0.5 ⫺0.2 2.0 3.1 0.1 ⫺0.2 4.4 Private transport 5.7
Producer prices
Staple grains ⫺0.5 0.1 3.4 2.6 0.1 ⫺0.3 5.4 Staple grains 7.1
Livestock ⫺1.4 0.2 2.5 2.3 ⫺0.3 ⫺0.3 3.0 Livestock 5.3
Other food ⫺1.3 0.8 5.0 1.7 0.5 ⫺0.7 6.1 Other food 6.3
Non-durables ⫺0.1 ⫺3.7 0.7 2.8 ⫺0.1 ⫺0.7 ⫺1.2 Nondurables 0.7
Durables 0.0 ⫺9.8 0.1 1.3 ⫺0.1 ⫺0.4 ⫺8.8 Durables ⫺9.7
Services ⫺0.2 0.8 1.3 2.4 0.0 ⫺0.1 4.2 Services 5.8
Margin services ⫺0.2 0.8 1.3 2.4 0.0 ⫺0.1 4.2 Margin services 5.8
Consumer prices
Staple grains ⫺0.4 0.2 3.1 2.5 0.1 ⫺0.3 5.2 Staple grains 6.8
Livestock ⫺1.2 0.3 2.3 2.3 0.1 ⫺0.6 3.2 Livestock 5.4
Other food ⫺1.1 0.8 4.5 1.8 0.2 ⫺0.3 5.9 Other food 6.2
Non-durables ⫺0.1 ⫺3.0 0.8 2.8 0.0 0.7 ⫺0.3 Non-durables. 1.6
Durables ⫺0.1 ⫺2.2 1.0 2.1 0.1 ⫺0.2 0.5 Durables 1.4
Services ⫺0.2 0.8 1.3 2.4 0.1 ⫺0.1 4.2 Services 5.8
Source: Hertel, Ivanic, Preckel et al. (2003a: table 5).

non-durables) fall as a consequence of liberalization by other LDCs

and tariff cuts, leading to a large change in the relative prices of food
and manufactures.
Turning now to the effects of these price changes on poverty,
Table 13.10 shows both the short-run and long-run consequences of
these price changes on the head–count ratio for each income group.
The total column indicates that poverty falls by 1.5 per cent in the
short run and 1.1 per cent in the long run. However, the changes are
not uniform across social groups. The increase in the relative price of
agricultural goods causes a sharp decline in poverty (2.8 per cent) in
poverty among the group deriving its income from agricultural
goods. By contrast the poverty headcount among the non-agriculture
group (perhaps this could represent the urban poor) actually
increases. The table indicates that a major cause of this is the liber-
alization of agriculture by developed countries, the effect of which is
Table 13.10. Effect of global trade liberalization on poverty in Indonesia (% change in headcount
across social strata grouped by primary income source)
Primary income source
Short-run Agriculture Non- Labor Transfer Diverse Total

effects of agriculture
Own-country of agricultural 1.3 ⫺0.9 ⫺0.7 ⫺0.3 0.3 0.5

liberalization goods
of non-agricultural 1.7 0.7 ⫺0.4 0.3 10.9 1.1
goods
Liberalization of agricultural ⫺4.1 3.5 2.9 1.6 ⫺0.7 ⫺1.1
by DC’s goods
of non-agricultural 0.1 ⫺1.5 ⫺2.4 ⫺1.0 ⫺0.8 ⫺0.7
goods
Liberalization of agricultural ⫺1.7 0.4 0.3 0.0 ⫺0.7 ⫺0.8
by LDC’s goods
of non-agricultural ⫺0.1 ⫺0.3 ⫺1.3 ⫺0.4 ⫺0.4 ⫺0.3
goods
Total short-run ⫺2.8 1.8 ⫺1.6 ⫺0.2 ⫺1.4 ⫺1.5
change in poverty
Long-run ⫺0.9 ⫺1.1 ⫺1.5 ⫺0.3 ⫺1.2 ⫺1.1
change in poverty
Source: Hertel, Ivanic, Preckel et al. (2003a: table 6).

to increase poverty by 3.5 per cent as a consequence of the relative

price changes described above. For Indonesia, the population in the
two income categories that lose (transfer and non-agricultural)
makes up just 14 per cent of the poor. Consequently national poverty
falls as a result of global trade liberalization in Indonesia.
Table 13.11 shows Hertel, Ivanic, Preckel et al.’s (2003b) findings
for several countries. They suggest that multilateral trade liberaliza-
tion will increase the poverty headcount in Thailand, Peru, and
Venezuela and reduce it in all the other countries except Zambia,
which experiences no change. In Brazil, for example, poverty rises for
the non-agricultural and labor strata, which together accounts for
more than 45 per cent of the poverty headcount, but poverty falls
amongst the agriculture group, which comprises 25 per cent of the
poor population. However, since the gains in agriculture are so large
a high proportion of the group leaves poverty, causing an overall
reduction in national poverty.
While, for reasons already explained, computable general equilib-
rium models need to be used with considerable caution, the detailed
Table 13.11. Effect of global trade liberalization on poverty for 14 developing countries (% change
in number of poor, relative to total population)
Primary income source
Country Agriculture Non-agriculture Labor Transfers Diverse Total
Bangladesh ⫺0.4 ⫺0.4 0.1 ⫺0.1 ⫺0.1 ⫺0.1

Brazil ⫺11.4 2.9 1.2 0.1 ⫺2.2 ⫺2.4
Chile ⫺25.0 3.4 2.3 0.7 ⫺2.4 ⫺3.9
Colombia ⫺8.9 0.5 1.0 ⫺0.1 ⫺2.1 ⫺2.2
Indonesia ⫺2.8 1.8 ⫺1.6 0.2 ⫺1.4 ⫺1.5
Malawi ⫺2.6 0.4 ⫺0.7 ⫺0.2 ⫺2.5 ⫺2.0
Mexico 2.5 ⫺0.8 ⫺0.7 ⫺0.2 0.2 ⫺0.2
Peru 2.2 1.1 3.9 0.6 1.8 1.4
Philippines ⫺5.2 1.1 ⫺0.9 0.0 ⫺3.0 ⫺3.1
Thailand ⫺0.2 13.6 8.8 6.9 4.9 5.7
Uganda ⫺0.2 ⫺0.8 ⫺0.7 ⫺0.2 ⫺0.5 ⫺0.5
Venezuela ⫺9.1 0.6 0.8 0.1 0.5 0.3
Vietnam 10.9 ⫺16.0 ⫺11.1 ⫺3.8 ⫺6.1 ⫺5.6
Zambia 0.0 ⫺0.1 0.2 0.0 ⫺0.1 0.0
Source: Hertel, Ivanic, Preckel et al. (2003b: table 8).

analysis provided by Hertel et al. highlights the differential effects

on different groups, and the fact that, while some individuals may be
lifted out of poverty, others will be forced into it. Certainly, the
distributional impacts cannot be ignored.
Policies to minimize the costs of adjustment
Trade liberalization can contribute to increased economic growth in

the long run. However, in the short run some social groups in devel-
oping countries may be negatively affected by changes in the prices
of the goods they consume and produce. Trade reform must therefore
be designed in conjunction with a range of complementary polices to
protect vulnerable social groups.
This section provides a brief survey of some of the main policies
that have been proposed to mitigate the effects of adjustment on
Social safety nets and credit markets

Even if the adjustment costs are quite small and short-lived, the
extremely poor in many LDCs may be incapable of sustaining them-
selves for short periods because of a lack of savings and the unavail-
ability of credit and insurance. For this reason one of the most
important components of trade reform is an effective social safety net.
Workers in industries which experience a negative shock through
lower foreign demand from lost exports, or lower domestic demand
as a result of increased competition from imports, may require fund-
ing if they lose their jobs in the adjustment process. Unemployment
benefits can enhance adjustment by giving workers the funds neces-
sary to search for alternative employment in different industries or
locations. Many developed countries already have comprehensive
safety nets, but developing countries will require assistance.24
24
Such assistance can also increase economic efficiency, by allowing workers to continue searching until
they find a job which better matches their skills.
14
12
10
8
6
4
2
0
Sub-Saharan South Asia East Asia and Latin America Middle East and Eastern and North America Western
Africa Pacific and Carribean North Africa Central Europe Europe
Figure 13.6. Expenditure on social security and welfare (% of GDP)

Source: Besley, Burgess, and Rasul (2001).
Figure 13.6 shows the average social security expenditure for

several regions. The poorest countries are not able to spend enough
to make their programs effective and will require international
assistance to meet the adjustment needs of trade reform.
Firms may also require assistance. Companies may need to make
adjustment-related investment in order to cope with new market
forces. In developing countries where capital markets are less
sophisticated, firms may be credit-constrained25 even if they would
be able to pay back the loans. The World Bank (1997) has reported
that lack of access to finance for new investments was the most
severe constraint small firms in Ghana faced after trade reforms in
1983. Changes in relative prices resulting from liberalization will
have short-run effects on cash flows that will differ across firms,
with the result that even if on average these cash flows improve, the
adverse effects on the losers can more than offset the positive effects
on the winners (See Greenwald and Stiglitz 1993).
Technical assistance
The Uruguay Round Decision on Measures in Favor of Least
Developed Countries called for ‘substantially increased technical
assistance in the development, strengthening and diversification of
25
Even in developed countries, small and medium-sized businesses often face severe credit constraints.
their production and export bases including those of services, as well

as in trade promotion’.
If the global gains from trade liberalization are as large as some
researchers suggest—the World Bank estimates that further liberal-
ization could yield an increase in real income by 2015 of more than
US$500 billion26—then it is reasonable to enshrine a principle of
compensation whereby those countries that suffer significant
adjustment costs relative to welfare gains should receive offsetting
assistance.
A principle of compensation is important for at least two reasons.
First, if the ‘development focus’ of the Doha Round is to have any
meaning, then WTO members must be mindful of the fact that the
cost of adjusting to their agreements will have serious consequences
for development. Not only do adjustment costs fall particularly
harshly on the poorest people in the world because they are least able
to afford them, but the costs also consume resources that would
otherwise be spent on alternative development priorities. For many
people, the impact of trade reform will overwhelm the effects of
other economic development programs.
The second motivation for the provision of compensation for adjust-
ment costs is the pragmatic need to win political support for reform.
High adjustment costs give some groups a vested interest in the status
quo. Identifying and compensating those groups may be an effective way
of removing impediments to welfare-improving global policy changes.
The purpose of technical assistance is to improve the trade perfor-
mance of developing countries through policy and strengthening of
institutions. A systematic review of technical assistance efforts is
beyond the scope of this chapter. In this section we note some of the
trends and limitations of existing programs and the need for more
wide-ranging support.
The responsibility for technical assistance has fallen largely on
international organizations. Both the World Bank and the WTO have
increased their technical cooperation activities. However, as much
26
The World Bank estimates that further liberalization of trade can generate up to US$500bn in static and
dynamic gains by 2015 (World Bank 2003). These estimates assume the elimination of agricultural export sub-
sidies and domestic support, a tariff ceiling of 10% for agricultural products and 5% for manufacturing in OECD
countries, and a 15% ceiling for agricultural products and 10% for manufacturing in developing countries.
as 90 per cent of financing for these activities comes from trust funds
provided by two or three bilateral donors, while the WTO itself has
typically allocated for technical cooperation activities less than one
per cent of its total annual budget—less than half a million US
dollars (see Michalopoulos 2000).
Trade-related technical assistance is often provided to assist
governments to implement existing agreements. This assistance is
often conceived by the provider and ‘supply-driven’ and is not related
to the overall priorities of the beneficiary (Prowse 2000). In addition,
technical assistance needs to be pro-active. It should strengthen the
recipient country’s ability to determine its own development priori-
ties and influence the outcome of WTO agreements.
A third useful expansion of technical assistance would extend its
scope towards ensuring that developing countries have access to
equal protection under the WTO’s dispute settlement system. Lack
of institutional capacity limits developing countries’ ability to pre-
sent and defend cases in the dispute systems, making those systems
manifestly unfair in practice. Developing countries are disadvan-
taged in complex and expensive legal proceedings. An expansion of
existing legal assistance schemes will be an important prerequisite
for institutional fairness.
The WTO Singapore Ministerial Conference in 1996 mandated a
more ‘integrated approach to assisting LDCs to enhance their trad-
ing opportunities’. In 1997, the Integrated Framework for Trade-
related Technical Assistance (IF) was launched with a view to
building trade capacity in developing countries (see Table 13.12).
The IF attempts to pull together the resources of several interna-
tional agencies to increase the scope and value of trade-related tech-
nical assistance. It also attempts to redress some of the common
criticisms of such assistance by ensuring that such assistance is
demand-driven, that it matches the specific needs of each LDC, and
that it enhances rather than undermines each LDC’s ownership of
trade-related technical assistance (UNCTAD 2002). Trade-related
technical assistance activities are broadly defined as:
● enhancing government institutions to manage trade policies

● assistance to create supportive trade-related regulations and policy
● strengthening export supply capabilities

● strengthening trade support and trade facilitation capabilities
Table 13.12. Trade-related assistance provided by multilateral agencies
Organization Activities
IMF Trade policy advice provided in the context of country surveillance and/or
program support, and considered in a broader economic and social
framework. Normally will include an assessment of the key complementary
policy requirements to support in-country trade reform— notably in fiscal policy
and the adequacy of social safety nets. Trade-related technical assistance
focused primarily on trade facilitation issues (customs administration) but also
on collation of data on external trade.
ITC Emphasis on enterprise-oriented aspects of trade policy and trade promotion
such as business implications of multilateral and regional agreements, private
sector involvement in trade policy, and management of regulation-related
issues by businesses. Hands-on training, assistance in data collation, analysis,
and institutional matters to favor private sector capability in trade
policy-making, managing of regulatory issues in trade, and compliance.
UNCTAD Policy analysis on trade and investment—advocacy of developing country
interests. Analysis of trade policy options in the context of economic
development. Trade-related technical assistance includes training and support
in trade negotiations and implementation of commitments, accessions advice,
and customs administration.
UNDP Trade policy options considered in the broader context of economic and social
development. Complementary policy analysis to support trade reform.
Sector-specific trade assistance in areas such as agriculture, fisheries, tourism,
and textiles. Private sector engagement in trade policy-making.
World Bank Trade issues are considered in a broader economic and social context of
development and investment-related policies. Creation and dissemination of a
core knowledge base that combines policy-relevant research, advocacy,
capacity-building, training, and operational support for trade at the country
level, including networking to link think tanks and trade policy makers within a
country.
WTO Emphasis on the WTO agreements. Factual information on WTO rights and
obligations of developing countries and progress in trade negotiations. Training
and consultation to assist developing country members in applying the WTO
agreements and using the WTO mechanisms.
Source: Prowse (2002).
In the initial phase of the implementation of the IF, the trade

assistance needs of forty LDCs were advanced. After limited success
several problems were identified with the IF approach. There had

been a substantial failure to put trade-related development issues at
the centre of national, agency, and donor priorities. At the country
level, national needs were selected without sufficiently broad con-
sultation. At the donor and agency level, insufficient attention was
given to integrating trade issues into the wider development agenda.
In 2001 an enhanced IF program was adopted with a view to
embedding trade-related capacity building into countries’ overall
development strategies through their national Poverty Reduction
Strategy Papers (PRSPs). For this purpose, ‘trade integration studies’
were commenced for a group of pilot countries. This strategy has the
distinct advantage of increasing the level of ‘ownership’ by LDCs.
The above list of trade-related assistance activities, however, is
inadequate in scope compared to what is needed. For instance, mon-
etary policies or structural adjustment programs advocated by the
IMF may adversely affect the flow or affordability of finance to facil-
itate the restructuring of the economy in response to liberalization.
While much of the technical assistance is designed to enhance the
ability of countries to design their own programs aimed at adapting
and responding to a new trade agreement, conditionalities associ-
ated with various forms of financial assistance may give them less
scope for doing so. Moreover, advice concerning how to cope with
the reduction of tariff tax revenues arguably reflects an inadequate
understanding of the nature of developing countries (e.g. the import-
ance of the hard-to-tax informal sector) and, as a result, leads to tax
structures which have adverse effects on growth and development.
Capturing the benefits of

liberalization for LDCs
Market access on its own is not sufficient to bring the benefits of

trade to developing countries. The UN Secretary General noted in
response to the European Union’s ‘Everything but Arms’ initiative
that ‘the LDCs have neither the surplus of exportable products nor
the production capacity to take immediate advantage of new trade

opportunities. They will need substantial investment and technical
assistance in order to expand their production’.27 Certainly, the lim-
ited increase in exports in the affected commodities from the least
developed countries to Europe is consistent with these concerns.
There can be no doubt among WTO members that tariff reduc-
tions must be accompanied by concerted efforts to ensure that poor
producers are able to capitalize on new trading opportunities. In par-
ticular, the Development Round faces the challenge of dealing with
two of the largest impediments to LDC export growth: supply con-
straints and product standards.
Supply constraints
Increased market access might generate a disappointing supply
response from many LDCs. In the context of low productive capacity,
a deficient policy environment, poor infrastructure, poor access to
technology, and missing/imperfect markets (especially financial
markets), liberalized markets will not stimulate the required devel-
opment to take advantage of new trading opportunities.
There has been some attention given to this issue within the WTO.
The final Declaration of the WTO Doha Ministerial meeting—which
was warned by the G77 countries about the lack of technical
assistance in recent years—reiterates the importance of technical
assistance and ‘reaffirms . . . the important role of sustainably
financed technical assistance and capacity-building programmes’
(para. 41).
Easing supply constraints requires a broader interpretation of the
responsibilities covered by technical assistance, i.e. more than bol-
stering public institutions. While public sector capacity-building is
an important objective, it is not a substitute for programs to enhance
the capacity of the private sector to develop into new markets.
A key component of private sector development is improved access
to finance—to take advantage of new opportunities for exports, there
27
Quoted in The Financial Times, 5 Mar. 2001.
must be export finance. In countries with underdeveloped financial

sectors, inadequate finance is a major constraint inhibiting exports.
To the extent that the poor are involved in trading activities, they
may face special difficulties in obtaining access to the trade credit
they need because of particular difficulties in assessing the credit-
worthiness of traders and because traders do not have sufficient
collateral.
Where there is an absence of private credit, there may be a role for
publicly funded institutions to increase access to finance for low-
income producers. For example, the Development Bank of Mauritius
(DBM) played a key role in providing finance for the expansion of
existing business and the establishment of new firms in Mauritius.
Among its several activities the DBM was involved in building
industrial estates to encourage development in export processing
zones (EPZs), setting up foreign exchange schemes for small and
medium-sized enterprises, providing working capital through
micro-credit, and extending preferential credit schemes.
Inadequate infrastructure is also an important source of supply
constraints. In particular, poor transport infrastructure can prevent
local farmers from getting access to large domestic markets and
international ports.
Another barrier to full participation in international trade is the
difficulty of establishing new industries in countries with poorly
diversified industrial bases. As noted earlier, when the EU intro-
duced its ‘Everything but Arms’ initiative in 2001, it extended duty-
free access to imports from LDCs in 919 product categories, but the
following year, imports were recorded in just 80 (Brenton 2003).
Failure to diversify is particularly evident in Africa, where the share
of agricultural value added in the GDP increased from 22 per cent to
25 per cent over the 1980–97 period while it fell from 18 per cent to
16 per cent over the same period for the developing countries as a
group.
Figure 13.7 traces the long-term trends in the commodity struc-
tures of Tanzania and Malaysia. Malaysia, like most South-East
Asian countries, was primarily agricultural in the 1960s and 1970s.
These countries pursued a successful pattern of industrialization
through import substitution followed by export-oriented growth. As
Tanzania 1961–2000 Malaysia 1961–2000

100 100
80 80
60 60
40 40
20 20
0 0
1961 1966 1971 1976 1981 1986 1991 1996 2000 1961 1966 1971 1976 1981 1986 1991 1996 2000
Broad primary industries Low-skill manufactures High-skill manufactures
Figure 13.7. Commodity structure of exports, Tanzania and Malaysia (% of

total exports)
Source: Bonaglia and Fukasaku (2002).
the ASEAN countries developed, they pursued a range of policy

measures to offset the anti-export bias resulting from local protec-
tion under the import-substitution policy. These policies included
incentives for private and foreign investment (particularly invest-
ment in new industries), EPZs, investment in infrastructure, and
duty drawbacks for exporting firms.
Product standards
As well as supply constraints, developing countries might suffer
from structural bottlenecks. Product standards—which require that
exported goods comply with a wide range of technical standards and
regulations set by the importing markets—are often a barrier to
developing country exporters. There are currently over 100,000
standards and technical rules in use around the world (UNIDO
2002). These standards are designed with the intention of facilitating
exchange and safeguarding health and safety. However, developing
countries may find compliance difficult or prohibitively expensive.
The potential for standards to have a detrimental effect on LDC
exports was recognized in the Uruguay Round. The Technical
Barriers to Trade (TBT) Agreement and the Agreement on the
Application of Sanitary and Phyto-sanitary Standards (SPS), both
negotiated as part of the Uruguay Round, were meant ‘to ensure that
technical regulations and standards do not create unnecessary obsta-
cles to trade’.28 Article 12.7 of the TBT Agreement specifically states:
‘Members shall . . . provide technical assistance to developing
country members to ensure that the preparation and application of
technical regulations, standards and conformity assessment
procedures do not create unnecessary obstacles to the expansion and
diversification of exports from developing country Members’.
In spite of these agreements many developing countries do not
have the ability to assist their producers to meet product standards.
There are serious deficiencies in infrastructure, processing
technologies, and national regulatory bodies. As a consequence,
significant assistance from developed countries is required to build
up their capabilities to conform to these product standard
requirements if trade liberalization is to have its intended impact on
the poorest countries.
UNIDO recommends a number of priority areas for international
assistance to the institutional development of developing countries,
including:
● A national/regional standards/standardization body: standards are
essential for production and trade, but also for consumer protec-
tion. To ensure that international (and national) standards are set
in a balanced manner, developing countries need to participate in
the drafting of such standards.
● A national/regional metrology system: a system that ensures that
the measurements and tests required for all production, quality,
and certification activities are consistent and correct. This
includes operational laboratories for primary and secondary phys-
ical standards as well as certified reference materials for chemical
and microbiological purposes.
● A certification/conformity assessment system: a system includ-
ing internationally recognized testing facilities that are able to test
products and certify that products and management/production
processes comply with applicable requirements and standards.
28
Preamble to the TBT Agreement.
● An accreditation system: a system which evaluates calibration

and testing laboratories and other bodies involved in certification
of products, systems, and processes, with a view to ensuring that
testing facilities and methodologies, and thereby the certification
activities, satisfy international standards.
Conclusion
Trade liberalization creates adjustment costs as resources move

from one sector to another. This chapter has described several
sources of adjustment costs (broadly defined) and concludes that
adjustment to a post-Doha trading regime will be disproportionately
costly and difficult for developing countries because of the loss of
preference margins, the loss of revenue from trade taxes, institutional
weaknesses including the absence of adequate safety nets, large
implementation costs, lack of the finance required to restructure
the economy, and the limited ability of poor populations to
manage short-term unemployment.
The effect of adjustment can be mitigated by effective national and
international policies to reduce the costs and facilitate the adjust-
ments. For instance, economies facing a new onslaught of imports as
a result of trade liberalization must find mechanisms to provide
credit for the creation of new enterprises and the expansion of exist-
ing enterprises to take advantage of the new export opportunities,
and macro-economic policies must be sensitive to these needs,
ensuring that (real) interest rates are kept appropriately low. In the
past, international institutions advising developing countries have
not been sufficiently sensitive to these needs. But more than good
policies will be required. There is a need for assistance, for instance
to develop the required physical and institutional infrastructure and
to provide compensation to alleviate the suffering of adversely
affected groups. This in turn will require a coordinated and well-
financed international effort. In the absence of a significant increase
in international assistance, responsibility for these policies will fall
on resource-constrained domestic governments, and trade reform (if

it is pursued at all) will come at the expense of other development
priorities. As a result, even a development-oriented round of liberal-
ization may fail to produce the growth benefits promised by the
advocates of a new trade agreement.
APPENDIX 1
Empirical review of market access issues
The ‘Doha Development Agenda’ of November 2001 puts poverty-reducing

economic growth at the center of the WTO’s considerations. If the develop-
ment focus of the Doha Round is to be a meaningful operating principle,
then the overriding task of the Round must be to ensure that the liberalization
agreements promote development in poor countries. In practice this means
giving priority to reforms which yield the largest benefits to developing coun-
tries, helping governments move towards good trade policies, and dealing
effectively with the implementation constraints faced by poor members.
A key theme of this book is that the WTO’s agenda should be driven by
economic analysis rather than the momentum of powerful interest groups.
This appendix supports the conclusions in the main part of the book by
reviewing the empirical evidence on the potential benefits and costs of
liberalization across various trade and factor flows. This type of analysis is
a crucial step to ensuring that priority is given to those elements of the
agenda that deliver the largest gains to developing countries.
This appendix is a brief survey of the effects of different liberalization
programs on global welfare. The bulk of useful regional-level empirical
studies use computable general equilibrium (CGE) models. Such studies are
based on simple models of the entire economy that, as the name suggests,
are developed in a computable form. These models enable us to observe the
effects of various liberalization experiments on trade volumes, prices, and
incomes. Simulations can separately determine the effects of reform on
different sectors and on different countries and regions. CGE models have
several limitations. They require a large amount of data (to estimate accu-
rately all the demand and supply functions which underlie them) and rely
on a few crude assumptions. Most importantly, they do not incorporate key
features of developing countries, such as the high level of unemployment.
Most assume away the problems posed by uncertainty, but the absence of
risk markets makes risk of central concern in developing countries. Most

assume perfect competition, while markets in developing countries may be
highly non-competitive. We present these models not because we believe
that they provide accurate assessments of the costs and benefits of trade
liberalization, but because they call attention to some of the key issues—
and because they have become a point of reference.
Where possible, the specific effects of reform on Commonwealth devel-
oping countries has been included. However, the empirical evidence in this
regard is thin and most global CGE studies do not disaggregate the effects on
small countries beyond the regional level.
This appendix analyses the potential gains and costs from liberalization
in four areas: agricultural trade, services, the temporary movement of
natural persons, and trade in manufactured goods.
To some extent the results of the survey give cause for a re-evaluation of
the current focus of negotiations. The estimated welfare gains from those
negotiating areas which attract considerable attention are estimated to
yield smaller benefits than other reforms on which there has been less
focus, and less progress.
Three conclusions which we believe are relatively robust emerge from
the empirical survey below:
● Liberalization of labor markets—in particular, allowing an increased
quota of workers from developing countries to work temporarily in
developed countries—offers large welfare gains.
● There are significant gains to be realized from the reductions in tariff
barriers to South–South trade. In both agriculture and manufacturing the
gains to developing countries from liberalization of trade between
themselves are estimated to be greater than those from liberalization of
trade with the OECD. (This may be both because tariff barriers between
developing countries are higher—implying greater benefits from
reductions—and because of the extensive use of non-tariff barriers by
developed countries.)
● There is considerable evidence that poorly implemented liberalization,
especially in the service sector, can have negative effects on the poor.
Carefully managed implementation, effective regulation, and substantial
assistance will be a necessary part of any reform agenda.
The literature surveyed identifies ambiguous effects of agricultural and

investment liberalization on developing countries. The reason for the
ambiguity will be detailed in the discussion below.
APPENDIX 1: REVIEW OF MARKET ACCESS ISSUES 217
A1.1 Agriculture
In the Doha Ministerial Declaration, WTO members committed them-

selves to reform of the main instruments of agricultural protection, includ-
ing ‘substantial improvements in market access; reductions of, with a view
to phasing out, all forms of export subsidies; and substantial reductions in
trade-distorting domestic support’. They also agreed that ‘special and
differential treatment of developing countries shall be an integral part of all
elements of the negotiations’.
This vision combines wide-ranging reform of the distorted agricultural
trade policies of developed countries and gradual liberalization in developing
countries.
This section surveys the (at times thin) empirical evidence on the
potential costs and benefits associated with the kind of reform envisioned
by the Doha Declaration. It focuses specifically on the welfare effects of
liberalization for developing countries.
Developing countries face the benefits of increased market access and the
(potential) costs of higher prices for domestic consumers. The fundamental
point is that consumers benefit from lower prices which result from large
agricultural subsidies, and producers lose. The net effect of wide-ranging
agricultural reform varies across developing countries depending on
the composition of their exports and imports of different commodities and
the price sensitivity of those commodities to liberalization. As a result, the
conclusions of the empirical evidence give cause for caution.
A reform agenda which maximizes the welfare of developing countries
must also recognize the specific effects of different protection instruments
on different commodities.
Furthermore, developed countries have a large number of instruments by
which they can redistribute income and alleviate poverty. In less developed
countries, by contrast, the set of instruments is far more circumscribed. Since
agriculture producers are among the poorest people in developing countries,
increasing the prices they receive may be one of the few instruments for alle-
viating rural poverty. But such policies are, at the same time, likely to
increase urban poverty. Tariffs on imports (especially when they countervail
subsidies by more advanced industrial countries), with some of the proceeds
used to provide targeted food subsidies, may accordingly increase welfare.
Reform should focus on maximizing market access benefits and identifying
ways of offsetting the terms-of trade-impact on consumers. This requires
(1) a rapid elimination of the most damaging protection instruments: export

and production subsidies on commodities that compete with developing
countries and which are not consumed extensively by developing countries,
or in which the effect of liberalization on prices paid by consumers in devel-
oping countries is likely to be small; (2) increased market access, particu-
larly for the goods exported by developing countries; (3) a gradual reduction
of production subsidies on sensitive commodities (those imported by devel-
oping countries with, say, substantial negative price effects on poverty); and
(4) assistance for the poorest countries.
While average manufacturing tariffs have fallen significantly in recent
decades, agricultural protection has remained stubbornly high. The average
level of agricultural producer support in OECD countries ranges from less
than 5 per cent of gross farm receipts in Australia to 20 per cent in the US
and 35 per cent in the EU (see Fig. A1.1). Developing countries face high
tariff barriers on many of their agricultural exports—the average tariff on
agricultural goods exported to developed countries was 15.1 per cent in 2000
(Hertel et al. 2000).1
80
1986–8 1999–2001
70
60
50
40
30
20
10
0
European Union
Slovak Republic
Czech Republic
United States
Switzerland
New Zealand
Australia
Hungary
Norway
Canada
Mexico
Poland
Ireland
Turkey
OECD
Korea
Japan
–10
Figure A1.1 Agricultural producer support, 1986–1988 and 1999–2001 (% of value of

gross farm receipts)
Source: Anderson (2003).
1
There are also large non-tariff barriers for some commodities, e.g. sugar quotas.
Table A1.1. Average protection in agriculture and food, 2005

This table shows the average protection (% ad valorem) for food and
agriculture by sector. The figures are worldwide averages in 2005. Subsidy
equivalents are aggregated across all regions and divided by exports at
domestic market prices.
Import tariff Export subsidy Production subsidy

(1) (2) (3)
Food grains 23 1 6
Feed grains 97 4 11
Oilseeds 4 0 9
Meat and livestock 17 8 2
Dairy 23 27 2
Other agriculture 11 0 0
Other food 1 0 0
Beverages and tobacco 18 0 0
Source: Hertel, Anderson, Francois et al. (2000: 26).
Table A1.1 shows the (projected) levels of farm support in 2005 after the
Uruguay Round agreements are fully implemented (Hertel, Anderson,
Francois et al. 2000). Tariffs are particularly high in the feed grains, dairy,
and food grains sectors. Column 2 shows that dairy products are the world’s
most subsidized exports, followed by meat and livestock. Producer
payments are highest for grains and oilseed sectors and lowest for meat,
livestock, and dairy (col. 3).
Average levels of subsidies (or protection) do not necessarily provide a
good indicator of how distorted the market is. For instance, most cotton pro-
duction may be totally unsubsidized, but America provides large subsidies,
which has a substantial marginal effect. The price effect of large subsidies to
even a limited group can be quite large, given the inelasticity of demand for
agricultural goods.
Table A1.2 shows the average tariffs on agricultural goods by importing
and exporting region. Developing countries face even higher tariffs on
exports to other developing countries (18.3 per cent) than on exports to
developed countries (15.1 per cent).
A second important determinant of the welfare effects of liberalization is
the agricultural trade balance across countries. Table A1.3 reports the
average trade specialization indices for several countries and regions over
the course of three decades. These indices give a measure of the trend of
agricultural trade balances through time. The value of the index ranges
Table A1.2. Average agricultural tariff rates (%)

This table reports the average tariff rates faced by high- and low-income
countries on their own and each other’s goods.
Exporting region Importing region
High-income Developing
countries countries
High-income countries 15.9 21.5

Developing countries 15.1 18.3
World 15.6 20.1
Source: Hertel and Martin (2000).
from ⫺1 for a country that imports and does not export a particular
commodity, to ⫹1 for a country which only exports it.
Table A1.3 shows a division between temperate products (program crops
and livestock), where developing countries are largely net importers and
developed countries are largely net exporters, and tropical products, for
which developing countries are largely net exporters. Many of the most
developed countries—the including the EU, the US, Australia, and New
Zealand—have increased their food trade balance over the last thirty years.
Most of the developing countries and regions—Sub-Saharan Africa, Latin
America (excluding Argentina and Brazil), Indonesia, Mexico, the Middle
East, and North Africa—have actually become more dependent on imports
in program crops and meat/livestock.
Table A1.4 gives an indication of developing countries’ trading relation-
ship with the developed world. It shows developing countries’ exports to,
and imports from, OECD countries as a share of each country’s total. As a
group, the OECD countries are big exporters of commodities to countries
like China, India, and the rest of South Asia (RSoAsia), and the Middle East
and North Africa (MENA). These countries are likely to be affected by lib-
eralization, which alters the price of OECD exports. Indonesia, Sub-Saharan
Africa (SSA) and China also rely on the OECD countries as export markets
for most of their products.
For countries like these, which are heavily integrated into OECD mar-
kets, liberalization brings risks and rewards. Producers gain from greater
market access, while consumers may lose through higher prices. These
effects are discussed in more detail in the next section. However, Table A1.3
gives us some indication of their relative importance across countries:
many developing countries are net importers of most commodities.
Table A1.3. Trade specialization indices, 1965–1998

The trade specialization index is calculated as (X ⫺ M) / (X ⫹ M), where X is exports and M is imports
for each country. A country that only exports has an index value 1. A country that only imports has an
index value ⫺1. Program commodities are composed of paddy rice, wheat, cereal grains, oilseeds,
raw sugar, processed rice, and refined sugar. ‘Livestock and meat’ includes livestock, wool, animal
products, meat, and dairy. ‘Other’ agriculture and food includes vegetables and fruits, plant-based
fibers, other crops, vegetable oils and fats, and other processed food.
n.a.: not available.
Years Program commodities Livestock and meat products Other agriculture and food
1965–75 1976–85 1986–98 1965–75 1976–85 1986–98 1965–75 1976–85 1986–98
Aus/NZ 0.95 0.97 0.94 0.99 0.98 0.98 0.13 0.10 0.32
Japan ⫺0.94 ⫺0.96 ⫺1.00 ⫺0.96 ⫺0.96 ⫺0.96 ⫺0.60 ⫺0.67 ⫺0.82
Korea ⫺0.90 ⫺0.82 ⫺0.90 ⫺0.14 ⫺0.73 ⫺0.85 ⫺0.23 ⫺0.23 ⫺0.21
USA 0.59 0.78 0.81 ⫺0.04 0.16 0.24 ⫺0.08 ⫺0.04 0.00
Canada 0.55 0.72 0.76 0.13 0.32 0.40 ⫺0.18 ⫺0.18 ⫺0.09
Mexico 0.19 ⫺0.87 ⫺0.83 0.03 ⫺0.41 ⫺0.54 0.66 0.56 0.36
EU15 ⫺0.74 ⫺0.56 ⫺0.27 ⫺0.49 ⫺0.05 0.13 ⫺0.48 ⫺0.37 ⫺0.17
EFTA ⫺0.91 ⫺0.89 ⫺0.76 ⫺0.08 ⫺0.02 ⫺0.04 ⫺0.27 ⫺0.27 ⫺0.08
CEU ⫺0.51 ⫺0.71 0.03 0.57 0.44 0.50 ⫺0.20 ⫺0.28 ⫺0.15
Turkey ⫺0.54 0.25 ⫺0.51 0.04 0.55 ⫺0.32 0.86 0.79 0.43
China ⫺0.17 ⫺0.55 ⫺0.18 0.87 0.69 0.38 0.22 0.36 0.28
Indonesia ⫺0.57 ⫺0.88 ⫺0.88 0.13 ⫺0.11 ⫺0.30 0.74 0.71 0.52
Vietnam n.a. ⫺0.37 0.85 n.a. ⫺0.65 ⫺0.01 n.a. ⫺0.10 0.48
ASEAN4 0.58 0.49 0.20 ⫺0.74 ⫺0.30 ⫺0.34 0.48 0.55 0.38
India ⫺0.58 ⫺0.15 0.43 ⫺0.40 ⫺0.24 ⫺0.10 0.43 0.24 0.44
RoSoAsia ⫺0.59 ⫺0.16 ⫺0.40 ⫺0.43 ⫺0.70 ⫺0.67 0.45 0.13 ⫺0.02
Argentina 0.97 0.99 0.96 0.99 0.92 0.75 0.64 0.71 0.78
Brazil 0.58 0.15 0.29 0.51 0.47 0.35 0.79 0.85 0.66
RLatAm 0.36 0.07 ⫺0.08 ⫺0.17 ⫺0.23 ⫺0.23 0.56 0.56 0.57
FSU n.a. n.a. ⫺0.63 n.a. n.a. -0.59 n.a. n.a. ⫺0.31
MENA ⫺0.91 ⫺0.97 ⫺0.94 ⫺0.80 ⫺0.94 ⫺0.87 ⫺0.01 ⫺0.54 ⫺0.45
Tanzania n.a. n.a. ⫺0.40 n.a. n.a. 0.18 n.a. n.a. 0.69
Zambia ⫺0.35 ⫺0.40 ⫺0.40 ⫺0.88 ⫺0.78 ⫺0.59 ⫺0.38 ⫺0.15 0.34
RoSSA 0.39 ⫺0.13 ⫺0.17 0.37 ⫺0.05 ⫺0.25 0.68 0.54 0.53
RoWorld ⫺0.10 ⫺0.43 ⫺0.66 ⫺0.27 ⫺0.50 ⫺0.45 ⫺0.16 ⫺0.25 ⫺0.43
Source: Dimaranan, Hertel, and Keeney (2003).
These results, however, have to be taken with some caution. They reflect
the pattern of trade flows that result given the highly distorted trade regime.
Many developing countries simply cannot compete against the huge subsi-
dies, say, to dairy. If these subsidies were not there, these countries would
become exporters rather than importers.
Table A1.4. Share of developing country trade with OECD, 1997 (%)
Commodity categories as for Table A1.3.
Program commodities Livestock and meat Other Agriculture

and food
Exports Imports Exports Imports Exports Imports
China 52 76 60 85 55 44
Indonesia 78 58 69 95 27 44
Vietnam 13 56 74 82 24 40
ASEAN4 40 48 54 71 47 44
India 27 75 52 85 31 24
RSoAsia 23 66 61 81 62 18
Argentina 23 58 38 35 57 36
Brazil 48 21 71 33 50 36
RLatAm 47 63 77 69 47 51
FSU 37 23 50 80 48 63
MENA 43 66 73 80 66 60
Tanzania 89 31 54 60 54 25
Zambia 86 7 69 93 65 43
RoSSA 63 49 77 82 69 62
RoWorld 62 73 59 66 62 61
Source: Dimaranan, Hertel, and Keeney (2003: table 4).
Moreover, agricultural markets are global markets, and so even if developing

country A imports wheat from developing country B, the price it pays is
greatly affected by the subsidies provided by the developed world or by trade
barriers that A might impose against the imports of the commodity.
The (national) real income effects of liberalization are dominated by two
factors: (1) the change in allocative efficiency, and (2) the change in terms-of
trade. Gains from allocative efficiency are realized when market distortions
are removed, permitting the economy to reallocate its resources to the most
productive use. These benefits accrue largely to the liberalizing region
itself. They are partially reflected in the huge budgetary savings that accrue
to the government of the liberalizing country.
The terms-of-trade effect comes from changes in a country’s export prices
relative to its import prices. The impact of global liberalization on national
terms of trade is usefully decomposed by McDougall (1993) into three
separate effects—the world price effect, the export price effect, and the
import price effect.
Export and import restrictions mean that there is a wedge between the
international price and the price that may be received by producers or paid
by consumers in each country. A country with an export subsidy faces a

higher producer and consumer price than the world price. A country with a
production subsidy faces a higher producer price (inclusive of the subsidy),
but the consumer price is the world price. A country with an import quota
or tariff faces both a higher consumer and higher producer price. Full liber-
alization entails eliminating all of these restrictions. Since there are huge
subsidies at present for the production of temperate agriculture products,
the world price of these crops would go up. The consumer price would go up
less in the EU,2 which has an export subsidy, than in the United States.
Developing countries that protect agriculture would see consumer and
producer prices go down relative to the world price, but since the world
price has gone up, the net effect is ambiguous.
Making matters more complicated are cross-commodity movements.
Not all agricultural crops are equally subsidized. The elimination of subsi-
dies would lead to a reallocation of resources within the agricultural sector.
It is possible, for instance, that with the elimination of the dairy subsidy, the
output of beef cattle could rise, and thus the price of beef might fall, even
though the consumer price of milk products might rise. In general, vegeta-
bles are less subsidized than grains, and hence there would be a shift away
from grains towards vegetables. Shipping costs are, however, less for grains
than for vegetables, and vegetable markets tend to be localized. Thus, even
if vegetable prices in developed countries fall, it will have little impact on
For a few crops, like sugar, there are quotas. The elimination of the quota
would lower producer and consumer prices in those developed countries
imposing quotas, and raise international producer and consumer prices.
The overall impact on developing countries depends on who receives the
quota rents. Even when the rent goes to those in developing countries, it
does not seep down to producers.
Differences between production and export subsidies are often exagger-
ated. A production subsidy of tp per cent in an economy producing xp raises
output by approximately tp␩s, where ␩s is the elasticity of supply. The
output impact per dollar of subsidy is tp␩s/tp px ⫽ ␩s/px, and the impact on
exports is ␩s/px␣, where ␣ is the ratio of exports to total production. On the
other hand, an export subsidy raises both the production price and the
consumption price, so that output is increased and consumption reduced.
The impact on exports is thus ␩s/px␣ ⫹ (1/␣ ⫺ 1)␩d/pc, where c is con-
sumption. While the impact on developing countries per dollar of subsidy is
2
Consumer prices could even fall.
greater, the difference is small if the elasticity of demand is small, which is

the case for many agricultural commodities.
It is important to recognize that among the ‘consumers’ of agricultural
goods is agro-business. While the higher price discourages direct consump-
tion, it also discourages agro-business, shifting it to other countries, includ-
ing developing countries. Thus, while there remains a presumption that
export subsidies are worse than production subsidies for developing coun-
tries, there is some question not only about the magnitude, but even about
the sign.
There are several additional concerns about the effects of agricultural lib-
eralization on developing countries. The effects of Uruguay Round liberal-
ization were noted at the Marrakech Meeting, where the Ministerial
Decision on ‘Measures Concerning the Possible Negative Effects of the
Reform Program on Least Developed and Net Food-Importing Developing
Countries’ addressed the need to provide adversely affected countries with
assistance.
The effect of liberalization on poverty is difficult to determine, largely for
the same reasons that we noted that there was an ambiguity in the impact
on developing countries as a whole: consumers lose while producers gain.
One problem is that the limited resources of small farmers could prevent
them from taking advantage of liberalized markets unless credit facilities
are improved.
One of the most significant effects of liberalization on the poor is felt
through changes in the price of food. Anderson, Dimaranan, Francois et al.
(2000) model a general equilibrium framework and find that full liberaliza-
tion of OECD farm policies would have a large effect on the volume of inter-
national food trade (a 50 per cent increase) but only a small effect on prices
(a 5 per cent increase on average). Beghin, Roland-Holst and vander
Mensbrugghe (2003) find that the price rise is larger for some commodities.
However, this does not include the effects of reform to non-farm trade and
so may misstate the effect in a multisector agreement. But even in this
study, the expected price increases are not large—the smallest increases
(about 4–6 per cent) are in the wheat, rice, and coarse grains sectors. The
effect of price increases on poverty is hard to generalize across developing
countries and within countries. This is because the relationship between
liberalization and poverty depends on the shares of household income from
different factors (land, labor, capital)—the prices of which will change. The
size of these changes depends on factor substitutability, factor intensities,
and factor mobility. The impact of price changes on welfare depends on the
relative shares of different goods in the production bundle. Additionally,
liberalization could have effects on net transfers, including increased

welfare, remittances from distant relatives, or changed taxation levels
(Anderson 2003).
Despite these difficulties, Anderson (2003) argues that since most of the
poorest people are net sellers of food (or at least sellers of agricultural labor)
liberalization would reduce poverty. Increases in the price of food would
stimulate production and increase the demand for unskilled farm labor.
Because unemployment (both open and disguised) is chronic in developing
countries, this in itself would have enormous benefits.
Another concern is food self-sufficiency. Some fear that cuts to protection
by OECD countries will lead to unaffordably high international food import
bills (see e.g. Sharma, Konondreas, and Greenfield 1996).
Table A1.5 shows the proportion of the population that is undernourished
in several developing countries. The last two columns show the value of
food imports as a percentage of total exports and total agricultural exports.
Interestingly, however, net food importer status (greater than 100 per cent in
the last column) is not highly correlated with the FAO’s category ‘Low-
income food-deficit country’ (LIFDC). Botswana, Jamaica, and Peru are all
net importers but are not classified as food-deficient. Also Côte d’Ivoire,
Malawi, and Kenya all import less than 20 per cent of their export volume,
but are classified as food-deficient.
An additional concern is that the liberalization of agricultural trade
would prevent countries from managing external price shocks. However,
Zwart and Blandford (1989) argue that liberalization could lead to less
volatile food prices, since trade can even out surpluses and deficits across
countries with heterogeneous production shocks.
But governments may want to intervene to stabilize either producer or
consumer prices, especially in developing countries, where means for risk
absorption are limited. Thus, initiatives at agriculture liberalization should
leave scope for developing countries to implement such stabilization
schemes.
A final concern is that many of the least developed countries already
receive preferential access to OECD country markets. Some of the benefi-
ciaries of these agreements are concerned that their advantages might be
eroded under a broader multilateral agreement.
However, there are several reasons why multilateralism should be
preferred. First, preferential agreements harm those countries that are not
in the agreement, including many which are very poor. Borrell (1999)
discusses this point in the context of the banana dispute of the 1990s, for
which one study showed that for every dollar of benefit to producers of
Table A1.5. Income category and food trade status

LI, LMI, and UMI refer to the World Bank classifications of low-income, lower-middle-income and
upper-middle-income countries; LDCs are least-developed countries, as recognized by the UN; LIFDCs
are low-income food-deficit countries, defined by the FAO as those countries with a GNP per capita less
than $1,445 in 2000 and which are net importers of food defined on a calorie basis; NFIDCs are net
food-importing developing countries, as defined by the WTO Committee on Agriculture.
n.a.: not available
Population under- Income/food trade Food Imports

nourished (%) status groupings
(% of total (% of agricultural
exports) exports)
Bangladesh 35 LDC LIFDC 21 829

Botswana 25 UMI NFIDC 14 256
Brazil 10 UMI 7 30
Costa Rica 5 UMI 6 19
Côte d’Ivoire 15 LI NFIDC LIFDC 9 17
Egypt 4 LMI NFIDC LIFDC 20 542
Fiji n.a. LMI 9 52
Guyana n.a. LMI 7 23
Honduras 21 LMI NFIDC LIFDC 13 48
India 24 LI LIFDC 5 42
Indonesia 6 LI LIFDC 6 56
Jamaica 9 LMI NFIDC 12 111
Kenya 44 LI NFIDC LIFDC 13 32
Malawi 33 LDC LIFDC 13 16
Morocco 7 LMI NFIDC LIFDC 12 146
Pakistan 19 LI NFIDC LIFDC 15 134
Peru 11 LMI NFIDC 14 152
Philippines 23 LMI LIFDC 6 123
Senegal 25 LDC LIFDC 26 357
Sri Lanka 23 LMI NFIDC LIFDC 12 68
Thailand 18 LMI 2 14
Uganda 21 LDC LIFDC 20 41
Zimbabwe 38 LI 5 13
Source: Anderson (2003: table 4).
bananas in ACP countries, the regime harmed non-ACP developing country

producers by a similar amount and reduced the welfare of EU consumers by
13 dollars. This type of scheme does not seem to be a very efficient means of
assisting banana producers in ACP countries, who could be directly
compensated by gains from the removal of the preference.
Second, if developing countries only sell part of their production in pref-
erential markets, then they are selling the rest of their output at lower than
Table A1.6. Welfare and efficiency gains expected from a 40% liberalization in agriculture, 2005
Column (1) reports the efficiency changes as a share of food and agricultural value added. AgrMkt40
(col. (5)) excludes reductions in production subsidies whereas Agr40 covers all forms of protection.
Agr40 experiment ratios Total EV by experiment (US$m)

(percentages)
Region Eff/$VA Eff/EV EV/Exp Agr40 AgrMkt40 Manufact40 BusFinSvces T&Tsvces

(1) (2) (3) (4) (5) (6) (7) (8)
NAmerica 9 11 0.035 3,401 1,436 3,310 4,517 52,532

WEurope 6 104 0.369 36,959 27,810 8,180 8,532 128,593
Aus/Nzl 6 ⫺12 0.377 1,786 1,348 207 209 8,421
Japan 6 120 0.253 12,552 13,461 6,607 2,564 33,358
China 6 1,067 0.012 172 753 22,593 826 8,710
Taiwan 4 143 0.060 265 295 3,288 83 6,072
Other NICs 3 115 0.333 2,672 2,996 5,270 612 23,228
Indonesia 2 1,183 0.002 6 26 792 270 1,474
Other SEAsia 2 101 0.465 1,931 1,247 2,631 393 11,092
India 1 137 0.200 1,058 927 3,084 19 3,989
Other SoAsia 1 118 0.852 1,176 1,181 1,645 9 2,213
Brazil 1 64 0.245 1,988 1,683 4,491 457 3,625
Other LatAm 1 48 0.360 3,055 2,366 1,449 652 8,611
Turkey 1 123 0.142 338 332 619 70 3,524
Other MENA 0 ⫺15 ⫺0.202 ⫺1,506 ⫺718 1,074 231 16,667
EIT 0 142 0.033 301 282 1,391 1,865 10,265
SoAfrCU 0 46 0.080 129 54 283 128 1,897
Other SSA 0 31 0.194 436 529 249 30 4,496
RoWorld ⫺1 115 0.741 2,601 2,611 2,399 137 3,798
World 69,320 58,619 69,564 21,604 332,565
Source: Hertel, Anderson, Francois et al. (2000: table 8).
normal prices because of the depressing effect of OECD protection on prices

in the rest of the international market.
The empirical results below come largely from CGE (computable general
equilibrium) models.
Hertel, Anderson, Francois et al. (2000) report simulation results from a
40 per cent liberalization of all types of protection (including production
and export subsidies). The impacts of these changes on real income are dom-
inated by efficiency and terms-of-trade effects.
There are significant gains from increases in allocative efficiency. The
first column in Table A1.6 reports the efficiency gains as a share of food and
agricultural value added. Large gains accrue to countries with the most
distorted policies, such as Europe, the US, and Japan. In Western Europe the
efficiency gains from liberalization amount to more than 8 per cent of the
entire sector’s value added.
Hertel, Anderson, Francois et al. (2000) add these efficiency gains to the
terms-of-trade effects to calculate a measure of welfare gains, the ‘equiva-
lent variation’ (EV)—which represents the amount of money that would
make consumers equally well off had there been no liberalization. The sec-
ond column in Table A1.6 shows the ratio of efficiency gains to total gain
(EV). Where this is greater (less) than 100 per cent the terms-of-trade effects
are negative (positive). For example, India experiences a terms-of-trade loss.
Sub-Saharan Africa, Brazil, and Latin America experience a terms-of-trade
gain because they are net exporters of food.
Table A1.7. Change in world trade volume from agricultural liberalization (%)
Change in world trade resulting from a 40% reduction in protection across 3 sectors: agriculture,
manufactured goods, and services. AgrMkt40 excludes reductions in production subsidies whereas
Agr40 covers all forms of protection.
AgrMkt40 Agr40 Manufac40 Services40
food grains 1.9 ⫺7.2 1.2 0.5

feed grains 4.1 1 0.7 0.5
oilseeds 0.6 5.8 0.1 0.3
meat & livestock 5.6 4.9 1.1 0.3
dairy ⫺6.7 ⫺6.9 0.1 0.7
other agriculture 8.3 8.1 0.9 0.4
other food 12.1 11.8 0.5 ⫺0.1
beverages & tobacco 27.5 27.6 0 0.8
extract 0 ⫺0.1 1.8 0.3
textiles 0.2 0.2 16.3 0.3
wear & apparel 0.7 0.4 22.3 0.6
wood & paper 0 0 3.6 0.4
chemical & universal 0 ⫺0.1 4.6 0.6
metals 0 0 5.5 0.4
autos 0.3 0.5 9.4 0.9
electronics 0.1 0.1 4.1 ⫺0.1
other manufactures 0.1 0.2 5.2 0.2
house utilities 0 0 0.1 1
trade & transport 0.5 0.5 1.5 59.8
construction 0.3 0.5 0.4 18.3
business & finance 0.1 0.1 0.5 10.8
government & service ⫺0.1 ⫺0.1 0.8 39.2

The total global welfare increase from a 40 per cent liberalization of

agricultural protection is US$70 billion in this study. The distribution of these
gains across countries is regressive. By far the largest absolute gains (col. (4))
accrue to developed countries, Western Europe, and Japan, who benefit from
the reduction in their own subsidies. However, col. (3) shows a measure of
relative welfare which accounts for the importance of food in GDP in each
region. Although the benefits of liberalization to Western Europe are large,
the food sector only represents 5 per cent of GDP. Column (3) shows the
total gain (EV) as a percentage of expenditure on food in that region. This is
one way of representing the benefits of liberalization relative to the import-
ance of agriculture in the economy. On this category, the largest gains are
realized in Other South Asia (non-India), Rest of the World (RoWorld), Other
South-East Asia (non-Indonesia), the Other NICs, and then Western Europe.
For comparison, Table A1.8 shows the results of a second CGE study by
Anderson, Dimaranan, Francois et al. (2001). They estimate that the total
welfare gain from the liberalization of all (i.e. 100 per cent) of agricultural
protection is US$164 billion.3 The Anderson study reports the impact of
liberalization by different regions on other regions. It concludes that the
farm policies of the OECD countries—after the Uruguay reforms have been
accounted for—reduce welfare in developing countries by US$11.6 billion.
Table A1.8. Welfare gains from global removal of trade barriers, 2005 (US$bn)
Liberalizing Benefiting Agriculture Other Textiles Other Total

region region and food primary & manufactures
clothing
High-income High-income 110.5 ⫺0.0 ⫺5.7 ⫺8.1 96.6

Low-income 11.6 0.1 9.0 22.3 43.1
Total 122.1 0.0 3.3 14.2 139.7
Low-income High-income 11.2 0.2 10.5 27.7 49.6
Low-income 31.4 2.5 3.6 27.6 65.1
Total 42.6 2.7 14.1 55.3 114.7
All countries High-income 121.7 0.1 4.8 19.6 146.2
Low-income 43.0 2.7 12.6 49.9 108.1
Total 164.7 2.8 17.4 69.5 254.3
Source: Anderson, Dimaranan, Francois et al. (2001).
3
This study is roughly consistent with Hertel, Anderson, Francois et al. (2000), whose predicted
US$70bn gain was based on a 40% reduction of barriers.
This is a small number in comparison to the gains realized by developing

countries as a result of liberalization in other developing countries (US$31.4
billion), and the gains realized by developed countries as a result of their
own liberalization (US$110.5 billion). It is also a small number in compari-
son to the gains predicted from liberalization in the temporary movement of
natural persons. The reason many developing countries do not gain more is
not difficult to understand given the structure of these models. Their gains
from more efficient resource allocation are offset by an adverse change in
the region’s terms-of-trade.
Moreover, these models simply add up the gains and losses. No note is
made of the fact that rural producers may be far poorer than the average
person in society; or that those who buy imported food (say wheat) are far
richer than those who live off locally grown crops, like millet. Nor does it
take into account any multiplier effects, e.g. the losses in income to
producers may have a larger effect on GDP than corresponding gains in
income to consumers (e.g. because of differences in marginal propensities to
consume). With unemployment rampant in most developing countries,
aggregate demand is often a constraining variable. Nor do these models take
into account the effects of credit constraints: higher prices allow poor rural
farmers to buy more fertilizer and higher-quality seed, thus providing a fur-
ther boost to their income. The higher incomes, in turn, may allow other
forms of high-return investment—including temporary migration into
higher-paying urban areas. Finally, there is considerable evidence that at
very low incomes, productivity depends on nutrition, and the higher
incomes accordingly will have a direct impact on agricultural productivity—
another effect not incorporated into these models.
For further comparison, we look at studies which focus on the effects of
specific protection mechanisms. Dimaranan, Hertel, and Keeney (2003)
examine the effect of domestic support (not including market access
restrictions) in industrialized countries on developing countries. The
terms-of-trade effect dominates welfare outcomes in their simulations,
leading them to conclude that a cut in OECD production subsidies would
lead to welfare losses in most developing regions. The first column in
Table A1.9 reports the average world price impacts of cutting domestic
support in all industrialized countries for all agricultural commodities by
50 per cent. The table shows that domestic support has the largest effect on
price for program crops (wheat, corn, barley, rice) and ruminant livestock
(beef). Sugar and dairy, which are mainly protected by tariffs, show small
price declines, and land and labor shifts out of program crops. The remaining
columns in Table A1.9 decompose the world price effects by type of domestic
Table A1.9. Change in average world prices due to comprehensive OECD domestic support
Commodity World price Contribution by tax/subsidy to world price

change change
Output Int.Input Land Capital
pdrice 0.26 0.12 0.34 0.05 ⫺0.23

wheat 4.91 1.03 1.68 1.11 1.09
crsgrns 5.5 1.42 1.79 1.02 1.27
oilsds 3.53 0.92 1.21 0.79 0.6
rawsgr ⫺0.58 0.09 0.14 ⫺0.33 ⫺0.48
othcrops ⫺1.5 ⫺0.01 ⫺0.03 ⫺0.69 ⫺0.77
ruminants 4.3 0.48 0.95 ⫺0.38 3.25
nonrumnts 0.54 0.26 0.45 ⫺0.14 ⫺0.02
rawmilk 0.21 0.14 0.81 ⫺0.33 ⫺0.4
pcrice 0.27 0.13 0.12 0.06 ⫺0.03
vegoilfat 0.97 0.2 0.34 0.24 0.2
refsgr ⫺0.06 0.05 0.06 ⫺0.03 ⫺0.15
rummeat 2.21 0.31 0.56 ⫺0.11 1.44
nrummeat 0.43 0.17 0.28 ⫺0.06 0.04
dairy ⫺0.19 0.14 0.36 ⫺0.27 ⫺0.43
othprfood 0.22 0.06 0.11 0.07 ⫺0.03
mnfc 0.12 0.01 0 0.1 0.01
srvc 0.11 0.01 0 0.1 ⫺0.01
instrument: output subsidies, intermediate input subsidies, land-based

payments, and capital subsidies (including livestock-based payments).
As shown in Table A1.10, the welfare impacts of domestic support reduc-
tion arise from allocative efficiency, output stimulus, and terms-of-trade
effects. The two largest agricultural exporters, Argentina and Brazil, gain
considerably in each of these categories. For these countries the terms-of-
trade effects are large and positive. However, for most developing countries
the terms-of-trade effects are negative and exceed the allocative efficiency
gains. A 50 per cent reduction in OECD domestic support results in a
decline in aggregate developing-country welfare of US$357 million.
Turning our attention to another experiment in partial liberalization,
Hertel, Anderson, Francois et al. (2000) analyse the effect of reductions in
border protection (leaving production subsidies unchanged). They report
that, not surprisingly, the global gains from this partial liberalization are
smaller than when liberalization also includes production subsidies: US$59
Table A1.10. Welfare impacts of domestic support reform ($m)
Region Equivalent variation Terms-of-trade components region
Total Allocative IS effect Terms of trade World price Export price Import
efficiency price
China ⫺69.1 ⫺69.6 ⫺18.0 18.5 ⫺51.8 137.1 ⫺66.8

Indonesia ⫺13.6 0.8 ⫺1.9 ⫺12.4 ⫺54.5 35.5 6.6
Vietnam ⫺8.2 ⫺1.9 0.3 ⫺6.6 ⫺10.0 5.8 ⫺2.4
ASEAN4 ⫺15.2 4.9 ⫺4.3 ⫺15.9 ⫺47.4 113.4 ⫺81.9
India 35.9 15.2 ⫺2.1 22.8 ⫺22.9 38.6 7.1
RoSoAsia ⫺44.2 ⫺3.3 ⫺1.2 ⫺39.7 ⫺57.2 17.2 0.3
Argentina 157.3 26.2 10.6 120.5 183.1 ⫺53.1 ⫺9.5
Brazil 200.2 73.3 31.9 94.9 1.1 88.5 5.3
RoLatAmer ⫺214.3 ⫺29.9 ⫺1.0 ⫺183.4 ⫺244.7 101.8 ⫺40.5
MENA ⫺270.1 ⫺50.6 ⫺1.8 ⫺217.7 ⫺315.9 83.1 15.1
Tanzania ⫺7.0 ⫺1.2 ⫺1.0 ⫺4.9 ⫺7.1 1.8 0.4
Zambia 0.0 0.2 0.0 ⫺0.3 ⫺1.4 0.4 0.7
RoSSA ⫺126.1 ⫺16.0 ⫺2.1 ⫺108.0 ⫺149.7 31.1 10.6
RoWorld 17.1 27.7 ⫺1.1 ⫺9.4 ⫺221.4 285.9 ⫺73.9
LDC total ⫺357.3 ⫺24.2 8.4 ⫺341.6 ⫺999.7 887.0 ⫺228.9
billion rather than US$70 billion (see Table A1.7). However, the additional
benefit from including production subsidies accrues entirely to developed
countries, which reap allocative efficiency gains. Western Europe alone
gains an additional US$9 billion when production subsidies are included. By
contrast, many developing countries are worse off when production subsi-
dies are liberalized because of the terms-of-trade effect. The Middle East and
North Africa and Sub-Saharan African countries outside the customs union
are significantly worse off as a result of the reduction in production
subsidies (Table A1.7).
The quantitative studies above indicate that the effect of agricultural lib-
eralization on developing countries is complex. Competing efficiency and
price effects have different effects across heterogenous countries.
For the reasons explained earlier, these results should not be taken too
seriously. The underlying assumptions of the computable general models
do not provide a good description of the economies of developing countries.
The results are highly sensitive to assumptions about elasticities and cross-
elasticities of supply and demand.
Still, there are three lessons that emerge from these admittedly highly
restrictive studies. The first is that in the process of liberalization, many
developing countries will find themselves worse off, especially urban workers.
But these adverse effects can be mitigated by adjustment assistance from
the more developed countries, which at the same time leaves the more
advanced industrial countries better off. This is because of the large alloca-
tive inefficiencies associated with the distorted patterns of production.
The second is that, while a true Development Round has to go well
beyond agriculture, it must include agriculture, for three reasons: it is too
important to some of the developing countries; there can be no principled
trade agreement without including agriculture; and, not surprisingly, as a
consequence, it has taken on enormous symbolic value.
Third, the potential for losses does not suggest that multilateral reform
should be abandoned. Instead, it suggests that to share benefits among all
countries, reform must accommodate differences across countries. The
case for liberalization is particularly strong for those commodities, like cot-
ton, the elimination of whose subsidies would have little direct bearing on
the standard of living of those in developing countries.
The empirical results surveyed above indicate that uniform elimination
of all agricultural protection would result in negative terms-of-trade effects
for many developing countries and sharp declines in farm incomes in Europe
and North America. The latter are in a position to bear the costs; the former
may not be. A reform agenda must carefully discriminate between liberal-
ization instruments. Such an agenda would have three key components.
First, a significant reduction in border protection in developed countries
(particularly the EU), including tariff cuts and the elimination of export sub-
sidies. Tariffs on the goods produced primarily by developing countries, as
well as those consumed primarily in developed countries, should be reduced
most rapidly. For example, the elimination of US and EU quotas and tariffs
on sugar and tropical products would increase the price received by devel-
oping world producers but only have a small effect on consumer prices in
Second, domestic production support for price-sensitive necessities that
are widely consumed in developing countries should be reduced gradually,
with some of the savings in developed country subsidy budgets being
directed at ameliorating the adjustment costs of those in the developing
world. Many developing countries in North Africa, Sub-Saharan Africa and
Latin America (though not Brazil, Argentina, or Mexico) rely on imports of
subsidized grains and oilseeds from OECD producers. The empirical evidence
reviewed above suggests that these countries are particularly exposed to
agricultural reforms which might increase the price of some commodities.
Third, domestic support should be shifted from market price support to
alternative payment systems. Reinstrumentation of protection in OECD
countries towards the least trade-distorting instruments (such as land-based

payments) is one possible means of compensating OECD farmers while
minimizing the impact on developing world producers. But many of the
so-called non-trade-distorting subsidies do in fact lead to increased
production, and too much has been made of the distinction between export
subsidies and production subsidies.
This type of program is similar to the thrust of the OECD in its ‘Positive
Reform Agenda’ for agriculture (OECD 2002) and is supported by a series of
recent research contributions.4
A1.2 Services
This section analyses the potential gains from the liberalization of services.
Services represent an increasingly large share of GDP in both developed and
developing countries—but a much larger share in developed countries.
Indeed, the area of focus of trade negotiations during the past fifty years,
manufacturing, is increasingly becoming the province of developing coun-
tries. It is natural, then, that the developed countries like the United States
should shift their focus towards liberalization of services.
The bulk of the empirical studies surveyed below suggests that the liber-
alization of services could yield significant welfare gains—much larger than
the gains from agricultural or manufactured goods. The estimates of global
gains are as high as US$400 billion. The estimates are large because protec-
tion levels are high in the service sector, and services make up a large (and
growing) share of world trade. Additionally, services are key inputs into the
production of almost all goods.
The enthusiasm in the cross-country empirical literature is tempered by
negative experiences at the national level. Opening up markets has been
4
Rae and Strutt (2003) use a CGE framework to compare the welfare gains of liberalization in border
measures and domestic support. They find that improved market access generates far greater trade and
welfare gains than reductions in domestic support. They conclude that the WTO should focus primarily on
achieving reductions in border restrictions and give a lower priority to the elimination of domestic support.
Hoeckman, Ng, and Olarreaga (2002) focus on the effect of OECD agricultural reform on developing
countries. They reach the same conclusion—that developing countries’ interests are better suited through
tariff cuts rather than cuts in domestic support. It is still the case, however, that the elimination of domestic
subsidies for certain commodities (like cotton) is likely to have a small effect on consumers in a developing
country, and a large benefit to producers. There are other crops for which this is also likely to be true. The
critical distortion differs markedly across commodities. Their results are highly dependent on assumptions
concerning demand and supply elasticities, and therefore results may differ markedly across commodities.
In the case of sugar, it is trade restrictions which dominate; in the case of wheat, it is almost surely production
and export subsidies which dominate.
accompanied at times by a reduction in competition, and an increase in

prices.5 In the case of financial services, there are even allegations that the
supply of credit to medium-sized and small domestic enterprises has been
reduced. Financial and capital market liberalization has been associated
with greater instability, not higher levels of economic growth. These
adverse consequences, in turn, help to explain the unhappiness of many
countries about efforts to force further opening up of the service sector.
It is easy to explain the discrepancy between the models and the outcomes.
It is partly that, deficient as the models used to analyse the consequences of
agricultural liberalization are, those in the area of services are far worse.
They model trade in services in exactly the same manner as they model
trade in goods, and thus the models have all the problems we noted earlier.
But in addition, there are several further problems. There are some formal
barriers to trade: the United States, for example, does not allow coastal
shipping by ships of another nationality. There are restrictions on media
ownership. Foreigners may not buy spectrum in the US, and if they cannot
buy spectrum they cannot provide broadcasting services except by selling
services to Americans who own stations. But many of the barriers to trade
are more subtle and are typically hard to quantify. Because the estimates of
government-created trade barriers are unreliable, so are the resulting esti-
mates of the benefits that would accrue from eliminating them. Worse still,
debates about liberalization in services do not center around discussions of
lowering the effective barrier from, say, 40 per cent to 20 per cent. Rather,
they center around particular measures, such as privatization and elimina-
tion of particular regulations. In each of these cases, the ramifications of the
particular measure extend well beyond the impacts on trade; in many cases,
these are incidental. Inevitably, then, debates about service sector liberal-
ization devolve into fundamental debates about national economic and
social policy. Should the media, for instance, be controlled by a few rich, for-
eign firms, who are able to use their wealth to control the flow of informa-
tion to the citizenry? This is an issue which is fundamental to the
functioning of democracy and should not be relegated to trade negotiators.
(At the same time, we should recognize that there are certain service sector
liberalizations which are little different in their impact than a standard
trade liberalization; such is the case for construction and maritime
5
For example, privatization of utilities—such as South Africa’s experience of granting its newly privatized
telecommunications utility Telekom a 5-year monopoly—can lead to inefficient services (OECD 2002).
Similarly the poor regulation of financial sectors across South-East Asia contributed to instability prior to the
crises of the late 1990s. Poor electricity deregulation has led to problems in many countries.
services—areas which were not included in the Uruguay negotiations and

which are of some concern to developing countries.)
Part of the reason that the standard models are unpersuasive arises from
the fact that they fail to take account of the highly differentiated nature of
services and the large ‘local information’ content. There are other ways in
which trade in services is obviously different from trade in goods. For the
most part, services have to be consumed at the point of production. In the
case of haircuts, the point is obvious. But the same is true for retail sales,
hotels, and electricity. Thus, the issue of trade in services in inextricably
linked (as already noted) to the movement of capital and labor. Without
these, there can be little trade in services.
But movements of labor and capital introduce a host of other considera-
tions, quite different from those associated with trade in goods. The issues
of labor—which are of vital concern to developing countries—are dealt with
in the next section. The central issue of concern for capital flows is investor
protections. What the investors would like, of course, is a world without
regulation or taxation, but that would compromise the general well-being of
the developing world. Indeed, most of the failures of liberalization have been
because of failures to put into place an adequate regulatory environment
(including one which ensures competition).
Economic theory, of course, says that, under certain idealized circum-
stances (e.g. constant returns to scale), full global efficiency can be gained
either through the free mobility of capital or the free mobility of labor. But
in the general case, equating the marginal returns to capital will not suffice
to equate the marginal returns to labor.
What do international firms provide when they provide services? Why
might an American company have a competitive advantage in financial
services over a domestic company? Presumably, this relates to knowledge
and information, e.g. about how to organize the provision of the services.
There are, in fact, a variety of ways besides direct investment by which
such services are sold. Hotels and restaurants issue franchises, which have
been enormously popular. Firms may contract out management services.
Of increasing concern in recent years in the United States is the problem
of ‘contracting out’ services through the Internet. This is one area in which
production of a service can occur at a place different from that where the ser-
vice is consumed. Such services are, in many ways, very much like traded
goods. It is of importance to developing countries that this nascent industry
not be impaired by the creation of new trade barriers by the developed
countries.
A further concern is that many service sector liberalizations increase

poverty, by increasing prices of essential services or reducing access for the
poor. Private firms may be less willing to engage in cross-subsidization of
market segments in poor and rural areas. Even if liberalization leads to
lower average costs through increased competitiveness and efficiency,
prices for some end-users may rise. Mosely (1999) estimates the impact of
financial liberalization on access to rural credit in four African countries—
Uganda, Kenya, Malawi, and Lesotho. The study found that liberalization
expanded credit where it was accompanied by innovative reforms with
regulation focused on rural access and poverty reduction. However, merely
privatizing government micro-credit agencies had a negative effect on rural
areas, as witnessed by the consequences of reform in Malawi.
There is also a concern that some service sector liberalizations, even if
they increase economic welfare, narrowly defined, have an adverse effect on
community life. Each individual in the community values having the local
store. The local store owner, like other local businessmen, provides essen-
tial services for the community. But these are services that are not ‘priced’.
It pays each consumer to buy the goods for the lowest price. Walmart is thus
able to drive the local store out of business. But there are fundamental
questions: ‘Is the community better off with the local businessman replaced
by a hired manager that is rotated in and out of the community in three
years’ time?’
The Services sector account for half of GDP in developing countries (60 per
cent in developed countries) and are some of the fastest-growing industries
in many economies. The performance of the services sector is critical to
growth. For example, the strength of a country’s financial sector is a deter-
minant of growth (see Levine 1997; Carlin and Mayer 2003). Well-managed
and well-regulated banks lead to an efficient transformation of savings to
investment, ensuring an appropriate allocation of resources. Similarly, effi-
cient business services reduce transactions costs, and telecommunications
capabilities are an important prerequisite for development in many sectors.
The GATS framework provides for four modes of service delivery:
● Mode 1, cross—border supply, which is analogous to trade in goods, arises

when a service crosses a national frontier, for example, the purchase of
software or insurance by a consumer from a supplier located abroad.
● Mode 2, consumption abroad, occurs when the consumer travels to the
territory of the service supplier, for example, to purchase tourism,
education, or health services.
(a) 1,000 1988 (b) 50 Services 46.4

911.4
900 1997 All industry
40 38.1
800
31.7
700 28.3
30
25.7
600
$bn
20
%
500
15.3
400
304.5 9.5 10.3
10
300
200 0
10.1
100 46.6 –2.2
1.2 0.6 0.8 7.2 9.4 8.9 –4.2
0 –10
Sub-Saharan South Latin East Asia OECD Sub-Saharan South Latin East Asia OECD
Africa Asia America and Africa Asia America and
and the Pacific and the Pacific
Caribbean Caribbean
Figure A1.2. Foreign direct investment in services

(a) Regional FDI stock
(b) Compound annual growth rate
Source: OECD (2002).
● Mode 3, commercial presence, involves foreign direct investment, for

example, when a foreign bank or telecommunications firm establishes a
branch or subsidiary in the territory of a country. Figure A1.2 indicates
that the stock of service-sector foreign direct investment (FDI) in
developing countries is small relative to OECD countries. However, as
Fig. A1.2b indicates, service, sector FDI is growing more rapidly in
developing countries, with the exception of Sub-Saharan Africa.
● Mode 4, movement of individuals, occurs when independent service
providers or employees of a multinational firm temporarily move to
another country.
These four elements of the General Agreement on Trade in Services (GATS)

encompass the movement of both capital and labor for services provision.
This broad approach enables countries to bargain to exploit their compara-
tive advantage. For example, developing countries might exchange greater
market access for capital for more fluid movement of unskilled workers to
developed countries. Figure A1.3 shows that the existing commitments have
been lopsided, with the development of Mode 4 proceeding most slowly.
International service transactions remain heavily protected in many
countries. Table A1.11 estimates the tariff-equivalent protection levels for
the construction services, business and finance, trade and transport, and
government services sectors across various countries and regions. These are
taken from estimates by Francois (1999) based on predictions of the level of
100
90
Number of commitments
80
70
60
50
40
30
20
10
0
DC LDC AC DC LDC AC DC LDC AC DC LDC AC
Mode 1 Mode 2 Mode 3 Mode 4
Figure A1.3. Full and partial market access commitments under GATS
DC ⫽ Developed countries: LDC ⫽ Developing and transition economies;
AC ⫽ Acceding countries
Source: OECD (2002).
bilateral services trade with the US. Discrepancies between the actual level
of bilateral trade (from US trade data) and the predicted level are assumed to
result from protection. The estimates for the trade and transport and the
government services sectors are taken from Hoekman (1995).
Table A1.11 indicates that impediments to trade are quite high in trade
and transport, government services, and construction and that barriers to
trade in services are much larger than barriers to trade in manufactures and
extraction industries. Findlay and McGuire (2003) report that impediments
to international services tend to fall as income rises, except in some profes-
sional service activities. This is indicated in Fig. A1.4.
One has to be careful, however, about interpreting these data, which refer
primarily to Mode 1 impediments. There are some service sectors—like
haircuts—that are typically small businesses. Without Mode 4 liberaliza-
tion, there is likely to be little cross-border sale of these services, even
though standard economic theory would suggest that there would be large
gains to trade. Services are highly individualized, often requiring large
amounts of localized information. Thus, even with no artificially created
barriers, individuals in one country may prefer to deal with those from their
own community; those from their own community may be able to provide
the services that individuals want. Local banks may have a competitive
advantage in knowing who the good borrowers are (an advantage which may
more than offset the problem of correlated risk). Large multinationals,
geared to providing services to those in advanced industrial countries, may
Table A1.11. Estimated average rates of protection by region and sector, 2005
Note that figures for services are tariff-equivalent rates. For example, in China, the figures below
suggest that import prices in the construction industry must be 41% above their free trade level to
explain the relatively low share of imports in this market.
Region Food Manufactures Services
Construction Business & Trade & Government

financial transport services
NAmerica 5 3 10 8 69 34
WEurope 8 1 18 9 84 40
Aus/Nzl 4 7 24 7 91 31
Japan 58 2 30 20 71 32
China 18 20 41 19 96 42
Taiwan 41 4 5 3 93 36
Other NICs 21 2 10 2 82 37
Indonesia 5 8 10 7 85 43
Other SEA 25 12 18 5 88 40
India 40 35 62 13 96 41
Other SoAsia 37 20 46 20 92 41
Brazil 4 16 57 36 71 44
Other LatAm 9 10 26 5 79 43
Turkey 31 6 46 20 92 40
Other MENA 15 14 10 4 92 40
EIT 12 9 52 18 71 35
SoAfrCU 8 8 42 16 58 26
Other SSA 13 9 11 0 94 43
RoWorld 76 33 46 20 97 38
Source: Hertel, Anderson, Francois et al. (2000: table 4). Original sources: Francois (1999), Hoekman (1995).
find it difficult to provide the services demanded in poor developing countries.

In short, there are reasons to believe that even apart from artificially created
barriers to trade in services, such trade might be more limited than trade in
manufactured goods. In that case, even though the service sector is today
larger than the manufacturing sector in developed countries, potential gains
from trade, and the reductions of trade barriers, may be more limited.
Moreover, one has to distinguish protection from the legitimate role of
government in imposing regulations that promote a variety of concerns of
general interest, even when such regulations have the effect of discouraging
foreign firms. For instance, affirmative action requirements might have this
effect, yet it is a legitimate objective of government policy to advance the
economic well-being of the disadvantaged.
South Asia (3)

East Asia and Pacific (5)
Sub-Saharan Africa/ Financial
Middle East and North Africa (17) services
Eastern Europe and Central Asia (3)
Latin America and Caribbean (18)
High-Income (26)
Countries
Sub-Saharan Africa/
Middle East and North Africa (42)
Eastern Europe and North Africa (42)
South Asia (5)
East Asia and Pacific (8) Telecoms
Latin America and Carribbean (21)
High-Income (21)
0.0 2.0 4.0 6.0 8.0 10.0

Degree of openness
Figure A1.4. Service sector openness by region: financial services and tele-
communications
The index for financial services captures the restrictions on new entry, foreign own-
ership, and capital mobility (IMF’s Annual Report on Exchange Arrangements and
Exchange Restrictions). The openness index for telecommunications captures the
degree of competition, restrictions on ownership, and the existence of an indepen-
dent regulator (ITU-World Bank database for 1998).
Sources: Mattoo, Rathindran, and Subramanian (2001); OECD (2002).
Inefficient service industries operate like a tax on an economy. Since ser-

vices are essential production inputs for most goods, the price and quality of
services provided to other producers have major impacts on the whole
economy. This is particularly true in key service industries such as telecom-
munications, transport, energy, and finance.
For this reason the majority of gains from effectively managed reform
accrues to the liberalizing region itself. Domestic firms benefit from access
to services at lower prices while consumers gain, and employees in most
service industries earn higher wages than in manufacturing (OECD 2002).
This poses two questions. First, if the benefits of liberalization are so great
to the liberalizing country, why does one need to include such liberalizations
within a trade agreement? Won’t countries have an incentive to do that on
their own? The answer traditionally put forward is that this is part of the
political economy of trade liberalization: one gives up something of value to
the other side (refusing at the same time to do something that would be of
even greater value to oneself) in order to extract a concession out of the
other side. The problem with this argument, in the context of services, is
that typically the advanced industrial country as a whole has relatively

little to gain, though particular firms in the developed country may gain a
large amount. Thus, in the area of services, one is pandering to special
interests. Pandering to special interests is not only bad economic policy, it
is dangerous, because after the foreign firm comes in, the company continues
to put pressure on its government to put pressure on the foreign government
to pass legislation or regulations that are to its benefit, to renegotiate con-
cessions when they prove unprofitable, or not to abrogate a contract even
when there is clear evidence that the contract was only entered into because
of corruption.
The reason that particular firms have much to gain, though not necessar-
ily countries as a whole, is associated with the very reason that there are
gains to trade in services: these arise not out of the standard differences asso-
ciated with differences in factor supplies (after all, most of the production
actually occurs in the purchasing country) but out of differential informa-
tion and knowledge, including organizational capacity. If that information
is widespread within a society, it is more likely that that information can
easily flow abroad, especially in our highly interconnected global economy.
Walmart, Toys-R-Us, and AIG have certain strengths that may not (or, in
some cases, may) be easily imitated. When hard bargaining by the US
allowed Toys-R-Us to open up in Japan, Japanese children benefited from
the cheaper toys, as did Chinese workers, as China’s sales of toys increased.
But America benefited only to the extent that Toys—R—Us profits
increased. American jobs were essentially unaffected.
The second, related issue is why these liberalizations should be part of an
international trade agreement. Such agreements should focus on areas
where there is a global public good being provided, e.g. through the setting
of standards or dealing with global externalities.
For agriculture and manufacturing, most models report results domi-
nated by two main effects—allocative efficiency gains and changes in terms
of trade. For services liberalization, movements of capital across borders
generate additional effects. First, foreign direct investment inflows and out-
flows can lead to an expansion or contraction in the capital stock located
within a region. Changes in capital endowments affect national output, but
since the capital is still owned by foreigners, the effect on GNP is less than
on GDP.
A second effect on income works through the rents earned on foreign
direct investments. Rents are created by barriers to services trade which fall
during liberalization. It is conceivable that service sector liberalization
could increase GDP but lower GNP, as domestic providers of services lose
their rents, foreign firms capture rents associated with their superior
knowledge and information.
Unlike agriculture, the modeling which focuses on the benefits resulting
from assumed gains in the efficiency of the provision of services suggests
that the main beneficiaries of reform in terms of both absolute and relative
welfare gains (as a percentage of GDP) are the developing countries.
Dee and Hanslow (2000) use a CGE model called FTAP, to report that the
total global gain from liberalizing all post-Uruguay trade restrictions on
services is US$130 billion.6 This amounts to half the total gain (US$260
billion) from total post-Uruguay liberalization—with the other half made up
of gains from agriculture (US$50 billion) and manufactures (US$80 billion).
These are the projected gains in real income about ten years after the liber-
alization has occurred. They include the gains from increased trade and
more efficient resource allocation.
In a similar exercise by Hertel (1999), world welfare gains were predicted
to be smaller than those in Dee and Hanslow (2000). Hertel predicts that the
gains from liberalization in agriculture, manufactures, and services are
US$164 billion, US$130 billion, and US$55 billion respectively. This varia-
tion is largely accounted for by differences in modelling assumptions. In
agriculture, Hertel assumes no effective Uruguay liberalization post-1995,
leaving much more to be done and more gains to be realized. In manufac-
tures, the difference is largely accounted for by differences in the base year
(Dee and Hanslow use 1995, Hertel 2005). Applying the 2005 base to the
FTAP model accounts for 90 per cent of the difference. In the service sector,
Hertel models only liberalization in the construction and business services
sector. He also does not include liberalization of FDI.
Brown, Deardorff, and Stern (2002) use a CGE model to calculate the
welfare effects of a 33 per cent reduction to barriers in the service sector.
Table A1.12 shows that they expect global welfare to rise by US$413
billion. All of the countries listed experience welfare gains as well as
increases in real wages and returns to capital. Developed countries experi-
ence large welfare gains—US$142 billion for the EU, US$131.4 billion for
the US, and US$57.9 billion for Japan. Brown, Deardorff, and Stern note
that their results—which are dependent on the accuracy of the size of
barriers they estimated indirectly from trade flow data—show that the
liberalization of services is likely to yield significantly larger gains than
other reforms.
6
Welfare gains are reported in standard CGE ‘real income’ terms. Real income is a measure of national
income deflated by an index of national prices.
Table A1.12. Welfare effects of service sector liberalization

Welfare effects of a 33% reduction in barriers to service trade on imports, exports, terms of trade,
welfare, real wages, and the return to capital.
Country Imports Exports Terms of Welfare Welfare Real wages Return to

(US$m) (US$m) trade (%) (%) (US$m) (%) capital (%)
Australia & 2,354.4 1,962.3 0.385 1.050 5,379.6 0.694 0.657

NewZealand
Canada 2,244.0 2,136.3 0.083 0.811 5,910.4 0.317 0.316
European 35,478.1 35,336.8 0.032 1.295 142,003.2 0.553 0.546
Union & EFTA
Japan 14,797.7 15,501.6 ⫺0.067 0.891 57,875.1 0.247 0.277
United States 32,467.7 32,231.5 ⫺0.033 1.448 131,426.8 0.524 0.534
India 919.2 803.9 0.212 0.552 2,321.6 0.170 0.204
Sri Lanka 121.7 99.1 0.335 1.202 200.4 0.881 0.507
Rest of 374.3 286.7 0.286 0.689 804.9 0.293 0.453
South Asia
China 5,660.3 6,210.9 ⫺0.128 1.320 11,959.1 0.840 0.603
Hong Kong 7,587.2 8,058.4 ⫺0.611 4.382 5,643.1 5.638 5.927
South Korea 4,842.2 5,002.5 ⫺0.102 1.339 7,619.5 0.913 0.956
Singapore 3,325.1 3,776.2 ⫺0.297 3.322 2,470.8 4.821 3.972
Indonesia 1,401.3 1,469.4 ⫺0.072 1.256 3,177.0 0.327 0.307
Malaysia 1,487.6 1,466.8 0.049 1.267 1,514.5 1.026 0.928
Philippines 1,986.7 2,195.0 ⫺0.462 2.342 2,067.1 1.739 1.622
Thailand 3,324.2 3,625.3 ⫺0.413 1.401 2,886.4 1.088 0.904
Mexico 863.1 809.1 0.110 0.878 3,099.3 0.204 0.195
Turkey 1,733.3 1,462.9 0.589 1.781 3,745.9 0.695 0.884
Central Europe 3,841.7 3,744.5 0.061 1.409 5,227.2 1.067 0.996
Central & 4,199.9 4,442.8 ⫺0.179 1.050 18,363.5 0.256 0.272
South America
Total 129,009.6 130,621.8 413,695.4
Source: Brown, Deardorff, and Stern (2002: table 4).
Hertel, Anderson, Francois et al. (2000) also compare the gains from
services liberalization to those for agriculture and manufactures. They
report that a 40 per cent cut in protection in the business services and
construction sectors yields a US$22 billion gain. Their estimate of the
potential gains in the trade and transport sectors is US$332 billion. The
trade and transport sectors represent a large share of global trade in services
and provide a significant flow on benefits to other sectors of the economy.
Table A1.13 shows the wide distribution of these gains across developed and
Table A1.13. Welfare and efficiency gains from liberalization of agriculture, manufacturing, and
services
Column (1) reports the efficiency changes due to a 40% liberalization in agriculture as a share of food
and agricultural value added. The second column reports the efficiency gain as a proportion of the
total gain in terms of equivalent variation (EV)— where this is larger than 100, the terms-of-trade effect
is negative. Column (3) reports the EV as a proportion of total expenditure. Columns (4)–(8) report
the EV for 5 different sector liberalization experiments.
Agr40 experiment Total EV by experiment ($m)

ratios (%)
Eff/$VA Eff/EV EV/Exp Agr40 AgrMkt40 Manuf40 BusFinSvces T&Tsvces

Region (1) (2) (3) (4) (5) (6) (7) (8)
NAmerica 9 11 0.035 3,401 1,436 3,310 4,517 52,532

WEurope 6 104 0.369 36,959 27,810 8,180 8,532 128,593
Aus/Nzl 6 ⫺12 0.377 1,786 1,348 207 209 8,421
Japan 6 120 0.253 12,552 13,461 6,607 2,564 33,358
China 6 1,067 0.012 172 753 22,593 826 8,710
Taiwan 4 143 0.060 265 295 3,288 83 6,072
Other NICs 3 115 0.333 2,672 2,996 5,270 612 23,228
Indonesia 2 1,183 0.002 6 26 792 270 1,474
Other SEAsia 2 101 0.465 1,931 1,247 2,631 393 11,092
India 1 137 0.200 1,058 927 3,084 19 3,989
Other SoAsia 1 118 0.852 1,176 1,181 1,645 9 2,213
Brazil 1 64 0.245 1,988 1,683 4,491 457 3,625
Other LatAm 1 48 0.360 3,055 2,366 1,449 652 8,611
Turkey 1 123 0.142 338 332 619 70 3,524
Other MENA 0 ⫺15 ⫺0.202 ⫺1,506 ⫺718 1,074 231 16,667
EIT 0 142 0.033 301 282 1,391 1,865 10,265
SoAfrCU 0 46 0.080 129 54 283 128 1,897
Other SSA 0 31 0.194 436 529 249 30 4,496
RoWorld ⫺1 115 0.741 2,601 2,611 2,399 137 3,798
World 69,320 58,619 69,564 21,604 332,565
Verikos and Zhang (2001) analyse the sectoral impacts of liberalization in

financial and communication services. They estimate that the gain from
each sector is US$24 billion. In both sectors, the majority of the gains come
from removing restrictions that discriminate against foreign firms.
These gains are not of course divided equally across all countries. In Dee
and Hanslow’s analysis, their US$133 billion gain accrues disproportionately
to developing countries. The service sectors in many developing countries
are projected to expand as their relatively large barriers to entry are removed.
For example, the service sector in China (which captures a large part of the
welfare gains) is projected to increase by a third when its large barriers to
entry are removed. This is predicated, of course, on the assumption that
China cannot obtain the requisite knowledge and information to improve
its service sector without opening itself more fully, an assumption which is
increasingly looking dubious.
Dee and Hanslow report results for only a small number of Commonwealth
developing countries. The gain to Malaysia from global service sector liber-
alization is US$1 billion—equivalent to 0.7 per cent of GDP. The gains to
the ‘rest of the world’ which includes smaller developing countries is US$23
billion or 0.8 per cent of GDP. Australia, China, Mexico, Chile and
Indonesia all gain more from tertiary liberalization than from primary and
secondary combined.
Service sector liberalization has the potential to deliver large welfare
gains to developed and developing countries. But the results of attempts to
estimate these benefits need to be taken with even more caution than
results in agriculture and manufacturing. The localized nature of the ser-
vices and the information that leads to success in its provisions means that
the elimination of government imposed barriers may not necessarily lead to
as much increase in trade as these models predict, and the gains in efficiency
may be partially offset as rents are transferred from domestic to foreign
producers. Some worry that financial service sector liberalization may even
lead to a reduction in national output, as the supply of credit to domestic
small and medium size enterprises is reduced. Empirical work estimating
these effects is limited, and most of the CGE models simply proceeds by
assuming that the production and sales of services is little different than
those of agriculture and manufactured goods. Thus to the litany of qualifi-
cations to the use of CGE models noted earlier, the additional ones noted
here mean that the results need to be taken with circumspection.
Since a large part of the gains from reform in the services sector accrue to
domestic policy reforms, it is not obvious why international negotiations
are necessary to achieve desirable outcomes. If the main gains could be
achieved unilaterally, then what is the utility of multilateral negotiations?
Matoo (2002) observes that many developing countries are in a situation
where their ability to implement reform is hindered by opposition from
domestic lobbies. In this context, it may be useful for some countries to
undertake reforms in the context of the momentum of broad international
negotiations.
There are however other areas where coordinated reciprocity could yield
significant gains in the context of multilateral negotiations. On one hand
developing countries have a number of developed country market access

interests, particularly in the area of access to labour markets (GATS Mode 4),
but also in construction and back-office business services. This suggests the
prospect for a deal based on access to labor markets in exchange for a
developed country demand such as greater commercial presence by foreign
service providers in developing countries (Matoo 2002). But the shape of
earlier service sector negotiations undermined this rationale; financial
sector liberalization, for instance, preceded liberalization in construction
and other service sectors that were of interest to developing countries.
Moreover there are several cases in which international cooperation may
be valuable. One example is competition policy: a permissible merger in
one jurisdiction may have a detrimental effect on competition in a smaller
foreign jurisdiction. Fink, Matoo and Rathindran (2001) suggest that the
GATS should require domestic competition law to consider the effect of
collusive agreements on foreign markets. Second they suggest that foreign
consumers should have the right to take action in foreign courts against cor-
porations that abuse their market power.
A1.3 Temporary migration
As we noted earlier, the GATS recognises four modes of service delivery.

The temporary movement of natural persons (TMNP) is known as Mode 4.
It is by far the smallest in terms of trade flows and the volume of scheduled
concessions recorded under the GATS (see Fig. A1.3).
The limited commitments that have been made refer to high-skill
personnel—business executives etc.—whose mobility is closely linked to
foreign direct investment and is an issue of interest to business lobbies in
developed countries. Thus far, Mode 4 has not progressed in a way that allows
developing countries to use their comparative advantage in low-and-medium
skill labor-intensive services.
The empirical studies surveyed below suggest that an expanded Mode 4
could generate enormous welfare gains. The temporary movement of less
skilled workers from developing countries (where they are in oversupply) to
developed countries (where they are relatively undersupplied) is estimated
to increase world welfare by hundreds of billions of dollars, even if the scale
of the labor flow is modest.
The movement of natural persons is usefully divided into three
categories.
Flows from developed to developing countries
This category represents highly skilled technical or managerial workers

who work in developing countries, either providing specialized services
such as consulting and legal advice or fulfilling senior management roles in
foreign-owned firms. This is a widespread practice which aids the manage-
ment of multinational firms and supplies useful skills to firms in developing
countries.
Skilled flows from developing to developed countries
The emigration of skilled workers from developing countries is actively

encouraged by developed countries and provides clear gains to them. Over
30 per cent of all doctors and nurses in the British health care system were
born outside the UK. The same is true for more than 12 per cent of academic
staff in British universities.
From the perspective of developing countries, this flow is better known as
the ‘brain drain’. The loss of skilled local workers deprives the country of
various economic and non-economic spillovers. The brain drain reduces
total output, diminishes the competence of domestic high-skill sectors,
and erodes the tax base. To the extent that these skilled workers are
complementary to other factors of production, such as unskilled labor, the
emigration of these skilled workers leads to lower incomes for these other
factors. Desai, Kapur, and McHale (2001) point out that the one million
Indians living in the United States account for just 0.1 per cent of India’s
population but earn the equivalent of 10 per cent of India’s national income.
On the other hand, the temporary emigration of skilled persons can bene-
fit developing countries in several important ways. First, the possibility of
temporary migration for skilled workers may increase the returns to educa-
tion in the source country, inducing more investment in human capital.
Commander, Kangasniemi, and Winters (2002) argue that this leads to an
increase in skilled workers in the domestic economy (even taking account
of those that migrate out) which partly offsets the direct effects of the brain
drain.
Second, remittances from workers in developed countries back to their
families are an additional benefit of migration (Massey, Araugo, Hugo et al.
1998). Remittances are an economically significant transfer for LDCs. In
2002, the Inter-American Development Bank reported that US$32 billion in
remittances was sent to the countries of Latin America and the Caribbean.
This was far greater than total development aid and only slightly less than
foreign direct investment (Ellerman 2003). (Though impressive in size, it is

worth being circumspect about the potential for remittances to generate
sustained development. Martin and Straubhaar (2001) argue that income
from remittances is potentially less valuable than income from newly
established local enterprises or export earnings because the domestic
spillovers may be smaller.)
Additionally there is evidence that national diasporas are an important
source of growth. For example, the 50 million Chinese living abroad have
been remarkably beneficial to the Chinese economy. They are a source of
business experience, network connections, and capital.
Unskilled flows from developing to developed countries
The movement of unskilled workers to developed countries offers the great-

est gain because it is associated with the largest difference between factor
prices and the largest scope for movement, measured as number of willing
people. It is also, however, the subject of the greatest concern in developed
countries.
Developed countries experience benefits and costs from unskilled migra-
tion. Foreign workers are an important source of labor in developed coun-
tries. London’s catering industry depends on migrants for 70 per cent of its
labor force and a large proportion of seasonal agricultural workers are for-
eign (Home Office 2001).
Opposition to unskilled labor flows comes from the fear that foreigners
displace local workers and contribute to unemployment.7 A study by the
British Home Office (2001) examines the widely held perception that
immigration is detrimental to native workers. It concludes that unskilled
workers often fill jobs in low-paid and insecure industries. In many cases
these are jobs that native workers are unwilling to accept. In these jobs,
foreign workers are filling labor market gaps rather than displacing native
workers. Where migrants move in to industries with unfilled vacancies,
their presence has little effect on either employment or wages of domestic
workers.
Even in industries where migrants are competing with domestic workers,
the effect is not much different than the impact of labor-intensive imports
of foreign goods on domestic manufacturing.
7
Such concerns are, of course, not consistent with the CGE models, which typically are based on full
employment.
In an early study, Hamilton and Whalley (1984) suggest that if labor were
free to move between countries sufficiently to equalize wages around the
world, world output would rise by more than 150 per cent.8
Even using the more conservative assumption that part of the cross-country
difference in wages reflects productivity differences which persist irrespective
of location—e.g. health and education—the gains are large. Winters,
Walmsley, Wang et al. (2002) assume that workers from poor countries are
naturally only one third as productive as workers in developed countries.
They estimate the gain from full labor mobility to be 70 per cent of
world GDP.
Obviously full labor mobility is an extreme and impractical assumption.
Winters (2000) estimates that even a relatively modest increase in labor
mobility would increase world welfare by US$300 billion. This study assumes
that 50 million additional workers from developing countries are permitted
to work in developed countries. Winters assumes that when workers move
from a low-to a high-wage country, they make up one quarter of the wage
gap, i.e. three quarters of observed wage gaps are due to persistent differ-
ences in productivity.
These rough estimates have been subsequently corroborated by Winters,
Walmsley, Wang et al. (2002) using a general equilibrium model. They find
that if developed countries allowed temporary workers from developing
countries to increase their workforces by 3 per cent (8 million skilled and 8.4
million unskilled workers), world welfare would increase by over US$150
billion. Winters et al. use the GTAP model and database developed by
Hertel (1997). They assume that temporary workers make up half the pro-
ductivity differences between their home country and their host country
when they move.
The initial residents of developing countries (which are labor-exporting)
gain most from the increase in migration. Their share of the total gain is
approximately US$80 billion, more than developed countries and a signifi-
cantly larger fraction of their income. The largest part of this increase
accrues to temporary migrants themselves. In several developing countries
Winters et al. find that the remaining residents of developing countries gen-
erally experience a loss in welfare. Despite increases in remittances and an
improvement in their terms of trade (as the fall in GDP reduces the supply
of their goods), the decrease in labor supply leads to a fall in the return to
other factors which outweighs these gains.9 However for many
8
Assuming an elasticity of substitution between factors of 1.
9
This result is obviously based on the assumption that the migrants are not unemployed unskilled work-
ers, and that there is not an equilibrium level of unemployment.
Commonwealth countries including India and those in South Asia and

South Africa, the welfare of permanent residents increases. For these under-
developed countries, the increase in remittances outweighs the decline in
labor and capital income. Remittance income increases the demand for
domestic goods and allows the real wages of both skilled and unskilled
workers to rise. The welfare of permanent residents in India and the rest of
South Asia increases by US$16 billion and US$350 million respectively. In
South Africa, the welfare of permanent residents increases by US$82 million,
while the welfare of temporary migrants increases by US$4.4 billion.
The substantial benefits estimated to be available to developing countries
from the liberalization of temporary migration for the unskilled—related to
the huge differences in wages in developed and less developed countries—
suggest that this is a promising area of reform. The global efficiency gains
too are probably an order of magnitude greater than those associated with
capital market liberalization, which has been the subject of so much atten-
tion. Not surprisingly, unskilled workers in developed countries have been
worried about the effect of the migration of unskilled labor on their wages,
and have so far been effective in limiting the extent of migration. (On the
other hand, business interests in developed countries have been successful
in allowing migration of skilled workers; this may be partially because
unskilled workers have thought that such workers are complements, and
thus they too will benefit. But such skilled labor migration is of ambiguous
benefit to the developing world.) Winters et al. suggest that one possible
way forward is to include existing foreign worker schemes under the GATS
by scheduling them and subsequently extending them. Many countries
already have short-term foreign worker schemes for low-skilled jobs in
agriculture, tourism, and construction.
A second approach is to focus on subcontracting schemes in future Mode 4
negotiations. Restricting the movement of people to existing employees of
incorporated firms avoids many of the problems of mobility for individual
workers (but also limits the scope of benefits). The pre-employment
guarantee ensures that the workers will arrive with a job and increases the
likelihood of their return after completion of the project. Where firms are
responsible for their staff, they can provide housing, health insurance, etc.
These services may reduce the costs of mobility for some workers.
However, there are also many disadvantages associated with subcon-
tracting. First, many service transactions are not appropriate for subcon-
tracting. Limiting mobility to transactions that are suitable for provision by
subcontractors obviously diminishes the potential gains from liberalization
of migration.
In spite of the huge barriers that developed countries have imposed on the
movement of unskilled labor, the economic force leading to such migration
is so great that large amounts still occur, in spite of the barriers, and the
developing countries have received large benefits, e.g. as a result of remit-
tances. The developing countries have an interest in facilitating the flow of
these remittances and in improving the rights and living conditions of the
migrants (many of whom are illegal). These issues would be high on an
agenda for a Development Round of trade negotiations centered on the con-
cerns of the developing countries.
A1.4 Manufacturing
The Doha Ministerial Conference agreed to launch tariff-cutting negotia-

tions on all non-agricultural products. The aim is to ‘reduce, or as appropri-
ate eliminate, tariffs, including the reduction or elimination of tariff peaks,
high tariffs, and tariff escalation, as well as non-tariff barriers, in particular
on products of export interest to developing countries’ (Doha Ministerial
Declaration, para. 16).
Significant progress on tariff reduction has been made in several sectors.
The empirical evidence below suggests that this reform has been accompa-
nied by an increase in the share of manufactured goods in world trade and in
the share of manufactured goods in developing country exports.
However, several studies suggest that the potential gains from the Doha
Round might be larger than those realized as a consequence of Uruguay
Round reform. This may be because the Uruguay Round was tilted against
the interests of developing countries. The reduction in tariff peaks on goods
of interest to developing countries and the reduction in protection on
South–South trade are promising areas for reform.
Figure A1.5 shows that over the same period there has been a shift in the
composition of global export towards manufactured products, while the
share of agricultural products has fallen.
While the average rate of agricultural protection in OECD countries has
risen in the last three decades, manufacturing protection levels have fallen.
Average tariffs on industrial goods imported into the OECD countries fell
from around 40 per cent in 1950 to 1.5 per cent in 1998 (Hertel 2000).
Figure A1.6 shows that in the last forty years, the share of agriculture in total
developing country exports has fallen from 45 per cent to less than 10 per cent,
while the share of manufactures has risen from 23 per cent to 79 per cent.
0.9
0.8
0.7
0.6
Manufacturing
0.5
Agriculture
0.4
0.3
0.2
0.1
0
1966 1970 1974 1978 1982 1986 1990 1994
Figure A1.5. Share of world exports, manufacturing and agriculture, 1965–1995

Source: Hertel (2000).
0.9
0.8
0.7
0.6
Manufacturing
0.5
Agriculture
%
0.4
0.3
0.2
0.1
0
1965 1973 1981 1989 1997 2005
Figure A1.6. Share of developing countries’ exports, manufacturing and agriculture,

1965–2005
South–North trade
Despite their export shift from agriculture to manufactures (Fig. A1.6)

and their increasing share of the world trade in manufactures (Fig. A1.7),
developing countries as a group are still net importers of manufactured
goods and net exporters of agricultural goods.
Figure A1.8 shows that developing countries have succeeded in increasing
the quantity of manufactured goods they provide to major developed
40
30
20
%
10
Agriculture
Manufacturing
0
1970 1975 1980 1985 1990 1995 2000
Figure A1.7. Developing countries’ share of world trade, 1970–2000

Source: World Bank (2002).
8
1980 1985 1990 1995
7
6
% of consumption
5
4
3
2
1
0
European union United States and Canada Japan
Figure A1.8. Selected developed country imports from all developing countries,
1980–1995
economies. However, even in the last period reported (1995) the shares were
low, ranging from just over 3 per cent in Japan to under 7 per cent in North
America.
Aggregate data hides the existence of tariff peaks, which may restrict
access to developing countries’ products. For example, in the processed food
sector, Canadian, Japanese, and EU tariffs on fully processed food are 42, 65,
and 24 per cent respectively. By contrast, the least processed products face
tariffs of 3, 35, and 15 per cent in the same countries. Partly because of these
trade restrictions, the penetration of developing country processed food has
been limited (World Bank 2002).
Table A1.14. Average manufacturing tariff rates

Average tariff rates faced by high- and low-income countries of their own
and each other’s goods.
Exporting region Importing region
High-income countries Developing countries
High-income 0.8 10.9

Developing 3.4 12.8
World 1.5 11.5
Source: Hertel and Martin (2000).
South–South trade
Notwithstanding tariff peaks, developing countries’ goods are subject to

much higher barriers in other developing countries than in OECD coun-
tries. Table A1.14 shows that developing countries face average manufac-
turing tariffs of just 3.4 per cent in developed countries but 12.8 per cent in
Figure A1.9 shows the average MFN tariff on manufacturing by importer
in 1995 and 2005. The highest tariffs are in developing countries, particu-
larly India, China, and Other South-East Asia.
Brown, Deardorff, and Stern (2002) use a CGE simulation model to test
the effects of trade liberalization in manufactures. In their model domestic
consumers respond to reductions in protection by purchasing more
imported goods. Industrial sectors in each country expand or contract
depending on whether their protection is reduced by more or less than in
other countries. Countries with larger than average tariff reductions experi-
ence a real depreciation of their currency to maintain a constant trade bal-
ance.10 Welfare in their model is determined by the effect of these changes
on allocative efficiency and each country’s terms of trade. The authors also
incorporate non-tariff barriers. These generate rents to the preferred
exporters, which are lost upon elimination. Thus the effect of liberalization
may not be positive for all exporters.
Brown et al. initially apply their model to the Uruguay Agreement on
manufactures. They estimate the welfare gains resulting from a scenario in
which all countries reduce their tariffs as per the Agreement. Table A1.15
10
This is the kind of effect which is often ignored in popular discussions of trade liberalization, but is
absolutely essential when attempting to appraise the true (general equilibrium) effects.
80
1995
2005
70
60
% ad valorem tariff
50
40
30
20
10
0
Ro dia
ld
rS a
sia
Ot er M il
a
Ot r L a
r S Am
a
Ind EIT
So sia
Au a
Tu l
y
tal
N iwan
he a
s
W pan
pe
Z
IC
z
rke
he hin
he en
fric
ric
Ot eric
he As
/N
or
Ot Bra
To
ro
oA
e
In
rN
Af
Ja
on
W
he at
oA
Ot C
Eu
Ot SE
Ta
Am
r
h
Figure A1.9. Average MFN tariff on manufactures, 1995 and 2005

shows that global welfare increases by US$56.5 billion and the gains are
shared across all countries. The largest welfare increases in absolute terms
accrue to the European Union (US$17.4 billion); however, large relative
gains—expressed as a share of GDP—accrue to the Rest of South Asia, the
Philippines, and Malaysia.
Brown et al. follow this simulation with an estimation of the welfare
gains available in the Doha Round (Table A1.16). They present the results of
a 33 per cent reduction in post-Uruguay Round tariffs. They estimate that
the potential gains from the Doha Round (US$163.4 billion from a 33 per
cent reduction) are significantly larger than those realized in the Uruguay
Round (Table A1.15). In particular, most developing countries gain signifi-
cantly more as a share of GDP than they did in the Uruguay Round. The
authors suggest that this may be because the Uruguay Round was tilted
against developing countries.
Hertel, Anderson, Francois et al. (2000) use the GTAP model of global
trade to make similar estimates about the welfare gains from a 40 per cent
liberalization of post-Uruguay tariffs on manufactures. They find that the
global gain is in the region of US$70 billion, roughly the same size as the gains
Table A1.15. Estimated welfare gains from the Uruguay reductions in manufactures tariffs
Change in Exports ($m) Terms of Welfare Real Return to

imports ($m) trade (%) (%) ($m) wage (%) capital (%)
Australia & 2,848.0 2,527.6 0.347 0.327 1,674.8 0.345 0.300

New Zealand
Canada 1,071.9 1,354.5 ⫺0.086 0.127 926.3 0.137 0.114
European Union 16,826.6 15,358.5 0.145 0.159 17,405.6 0.157 0.163
and EFTA
Japan 8,680.6 8,331.3 0.062 0.102 6,608.4 0.092 0.115
United States 12,426.0 13,459.3 ⫺0.133 0.123 11,187.1 0.124 0.122
India 2,585.3 3,628.9 ⫺2.099 0.446 1,875.4 0.316 0.577
Sri Lanka 98.8 106.3 ⫺0.193 0.558 93.0 0.507 0.608
Rest of 3,454.8 4,820.1 ⫺7.541 2.025 2,366.5 2.224 1.828
South Asia
China 3,112.6 1,917.7 0.456 0.305 2,762.2 0.347 0.271
Hong Kong 763.5 480.1 0.254 0.360 464.1 0.346 0.373
South Korea 2,858.6 2,733.2 0.068 0.422 2,403.3 0.409 0.435
Singapore 3,539.8 3,647.5 ⫺0.078 2.111 1,570.3 1.943 2.258
Indonesia 936.5 894.5 0.068 0.247 626.0 0.291 0.215
Malaysia 2,790.9 3,411.4 ⫺0.563 1.919 2,293.9 1.816 1.974
Philippines 2,452.6 3,102.1 ⫺1.989 1.917 1,691.7 1.853 1.964
Thailand 1,264.7 1,002.3 0.291 0.366 753.9 0.597 0.283
Mexico ⫺64.9 1.4 ⫺0.026 0.019 66.3 0.038 0.010
Turkey 319.3 253.9 0.143 0.123 259.1 0.122 0.124
Central Europe 1,871.7 1,846.1 0.020 0.294 1,091.2 0.311 0.270
Central & South 3,778.8 2,999.5 0.423 0.022 377.1 0.043 0.004
America
Total 71,616.2 71,876.4 56,496.0
Source: Brown, Deardorff, and Stern (2002).
they predict from agriculture. Table A1.17 shows that, with the exception of
Sub-Saharan Africa, developing countries gain more from the reduction of
manufacturing tariffs than they do from reduction of tariffs on agriculture.
A third study, Hertel (2000), using the same GTAP model, estimates
that the benefits of full (100 per cent) reduction in post-Uruguay manu-
facturing tariffs is a global gain of US$130 billion. Again, the author pre-
dicts that a large share of this will accrue to developing countries.
Figure A1.10 shows the developing country share of the gains from reform
in three different sectors (and from combined reform). Manufacturing is
the sector most benefited within developing countries (which gain over
Table A1.16. Estimated welfare gains from manufacturing liberalization in Doha Round
This table shows simulation results from a 33 per cent reduction in post Uruguay tariffs on manufactures.
Change in Exports Terms of Welfare Real Return to

imports (US$m) trade (%) (%) (US$m) wage (%) capital
(US$m) (%)
Australia & 3,720.7 3,457.2 0.267 0.545 2,790.6 0.508 0.515

New Zealand
Canada 1,996.0 2,097.3 ⫺0.013 0.347 2,526.2 0.216 0.251
European Union 23,184.8 22,840.3 0.050 0.358 39,273.0 0.190 0.199
and EFTA
Japan 19,071.4 15,817.0 0.548 0.696 45,190.9 0.234 0.304
United States 20,454.2 18,337.3 0.167 0.260 23,634.2 0.198 0.224
India 3,280.4 4,054.2 ⫺1.384 0.733 3,084.4 0.439 0.592
Sri Lanka 536.8 592.1 ⫺1.025 3.207 534.5 1.565 2.010
Rest of 1,892.0 2,018.4 ⫺0.604 1.895 2,214.7 0.889 1.025
South Asia
China 16,080.3 19,416.3 ⫺1.221 1.199 10,859.3 1.470 1.323
Hong Kong 3,182.8 1,840.3 1.246 1.444 1,859.1 0.947 0.647
South Korea 8,023.4 8,440.7 ⫺0.233 1.515 8,622.9 1.158 1.003
Singapore 4,382.9 4,161.8 0.131 2.276 1,692.5 2.481 2.611
Indonesia 2,362.7 2,336.0 0.053 0.835 2,113.3 0.645 0.447
Malaysia 4,242.8 4,805.2 ⫺0.488 2.555 3,055.1 2.896 2.812
Philippines 3,984.0 4,535.1 ⫺1.192 5.478 4,834.4 3.310 2.461
Thailand 3,406.1 3,970.1 ⫺0.675 0.873 1,798.6 1.664 0.972
Mexico 916.3 1,132.6 ⫺0.166 0.364 1,283.1 0.195 0.204
Turkey 1,421.0 1,558.6 ⫺0.335 0.827 1,740.3 0.349 0.272
Central Europe 3,866.3 4,366.4 ⫺0.428 0.734 2,724.2 0.816 0.722
Central & South 5,038.9 6,103.2 ⫺0.612 0.206 3,610.0 0.159 0.108
America
Total 131,043.7 131,880.0 163,441.4
Source: Brown, Deardorff, and Stern (2002).
70 per cent of the welfare and efficiency dividend). This persistent result
in the literature is derived from the fact that developing countries have
the highest tariffs on manufacturing goods and thus receive the largest
gains from removing the distortions. The studies do not separately analyse
‘consumer benefits’ (access to goods at lower prices) and ‘producer bene-
fits’ (the creation of new jobs as a result of access to markets abroad).
Another implication of the analysis is that the realization of these alloca-
tive efficiency gains will entail significant adjustment costs—a theme we
revisit in Appendix 2.
Table A1.17. Estimated welfare and efficiency gains from a 40%

liberalization in agriculture and manufacturing, 2005
Both columns report the benefits of reform in terms of equivalent
variation.
Total EV by experiment (US$m)

Region Agriculture Manufacturing
NAmerica 3,401 3,310

WEurope 36,959 8,180
Aus/Nzl 1,786 207
Japan 12,552 6,607
China 172 22,593
Taiwan 265 3,288
Other NICs 2,672 5,270
Indonesia 6 792
Other SEAsia 1,931 2,631
India 1,058 3,084
Other SoAsia 1,176 1,645
Brazil 1,988 4,491
Other LatAm 3,055 1,449
Turkey 338 619
Other MENA ⫺1,506 1,074
EIT 301 1,391
SoAfrCU 129 283
Other SSA 436 249
RoWorld 2,601 2,399
World 69,320 69,564
0.9
Efficiency
0.8 Welfare
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
Manufactures Agriculture Services Full
Figure A1.10. Share of post-Uruguay global liberalization gains accruing to developing

countries
A1.5 Conclusions
The purpose of the empirical survey in this appendix is to suggest a priorit-

ization of trade issues that will benefit developing countries. When negotiat-
ing parties lobby at the WTO, it is assumed that they do so in their own
self-interest. In the case of developing countries and their advocates, it is not
always clear what evidence they are using to determine how different
reforms will affect them.
The CGE results presented in this appendix are certainly not perfectly
reliable estimates of the welfare effects of various WTO proposals—indeed
the estimates vary quite widely between different studies. However, they
draw attention to the wide range of global effects of WTO proposals.
The results of our empirical survey suggest a different prioritization from
the current hierarchy of market access issues receiving attention in the
WTO. The evidence we have presented suggests that an alternative market
access liberalization agenda might focus on labor market access for
unskilled workers, unskilled labor services, market access for agricultural
goods exported by developing countries, and tariff peaks for manufactured
goods. The empirical work has paid less attention to the non-market access
issues like competition policy but, as we noted in the text, here too the agenda
that has been pursued, e.g. in the Singapore Issues, is markedly different from
that which reflects the interests and concerns of the developing world.
APPENDIX 2
Empirical Review of the Singapore Issues
The inclusion of a domestic regulatory agenda in WTO negotiations

represents a departure from the traditional ‘market access’ focus of the
GATT rounds. The national regulations embodied in the ‘Singapore Issues’
have become more prominent as the liberalization of traditional trade
protection instruments has highlighted the trade impact of remaining
differences in national regulatory regimes.
Efforts to harmonize national regulation have commenced in competition
law, FDI policy, transparency in government procurement, and trade facilita-
tion. It has been argued that these issues are not priorities for low-income
countries and should not form part of a so-called ‘Development Round’.
In particular, there is significant opposition from developing countries. In the
space of a month from early June 2003, 77 developing countries, including
over half the WTO membership, made public statements urging that
new negotiations should not be launched as part of the Doha Round
(CAFOD 2003).
Several developing countries see the Singapore Issues as incursions into
their national sovereignty that are not justified by the benefits they bring.
Multilateral regulatory disciplines threaten the specter of preventing
individual governments from pursuing development policies based on their
own national priorities and problems.
In addition there are concerns that the initiatives based on the Singapore
Issues may impose a large burden on the administrative capacity of develop-
ing countries. There are significant costs associated with both the creation
and enforcement of new regimes in competition policy, investment regula-
tions, and trade and customs procedures. Moreover, the required institutional
capacity and human expertise may not be available in developing countries.
In OECD countries these critical inputs developed gradually over a long
period of time. These considerations suggest that the payoff of requiring
WTO members to implement rapidly the Singapore proposals may not be
large relative to the costs. Furthermore, it suggests that any reform will
require significant technical assistance from developed countries.
Finally, there is broad concern that the Doha agenda may be overloaded.
The Doha Round has an ambitious work program involving multilateral
negotiations on many issues. The inclusion of complex and controversial
issues may slow progress on more fruitful initiatives.
As the debate on the Singapore Issues evolved, two other issues became
more apparent. The first is that many of the Singapore Issues involve a
detailed knowledge of complex public policy issues that goes well beyond
the competence of trade ministers to negotiate. The outcomes, accordingly,
might not be good even for the developed countries. There was a resonance
with what had happened in the intellectual property negotiations (TRIPS)
in the Uruguay Round, where both the Council of Economic Advisers and
the US Office of Science and Technology Policy raised serious concerns, to
which the US Trade Representatives paid little attention in the negotiations.
The issues that were being debated under ‘Competition’ did not attempt to
unify treatment of predatory pricing between domestic and foreign firms,
a natural part of any attempt at developing a uniform competition code.
The United States put highly controversial issues involving capital market
liberalization on the agenda of the discussion of investment, which the
IMF had revisited almost contemporaneously, raising questions about
the potential economic benefits and costs.
This brings us to the second concern: some of the proposals would
have actually been adverse to the development of the developing
countries. They went against the entire spirit of the Development Round.
Such was the case, for instance, with proposals for full capital market
liberalization.
In the context of these concerns, we consider four criteria for prioritizing
an issue in the current round of negotiations:
● Is the WTO justified by returns to international collective action that

are higher than returns to unilateral action, i.e. are there spillovers or
externalities which justify multilateralism?
● Are the benefits of the initiative large relative to other proposals? Are the
benefits shared between developing and developed countries?
● Are the costs of implementation small relative to the benefits of the
initiative?
● To what extent do multilateral commitments impede national development
strategies?
APPENDIX 2: REVIEW OF THE SINGAPORE ISSUES 263
Using these criteria, this appendix analyses the merits of including the
Singapore Issues in the Doha Development Round. The merits are certainly
not uniform across the four issues or even across the different initiatives
within each issue.
Significantly more work needs to be done to quantify the potential benefits
and costs flowing from the Singapore Issues. Indeed, the paucity of authori-
tative studies in this area is in itself a reason to advocate caution. Therefore
the conclusions of this appendix are preliminary.
Nonetheless, the empirical survey below suggests that the current focus
of the WTO’s regulatory harmonization agenda is misdirected in some
areas.
In the competition and investment arenas, the WTO should move away
from imposing uniformity on manifestly different countries and focus its
attention on areas where externalities generate returns to multilateralism.
In investment policy, reducing the ‘race to the bottom’ incentive war would
be a useful initiative. Similarly, competition policy initiatives should
include anti-trust action against cartels which raise prices for developing
countries, and a mechanism for analysing the global effects of merger
decisions in developed countries. In government procurement and trade
facilitation, progress should be made through national efforts aided by
technical assistance, rather than through imposing additional obligations
in the WTO.
A2.1 Investment
At the Singapore WTO Ministerial Meeting in 1996, members agreed to

form a working group to study the relationship between trade and invest-
ment. Since then, some developed countries have attempted to move
towards a negotiated investment agreement within the WTO.
The proponents of such an agreement seek internationally binding rules
that would minimize the conditions and regulations on foreign investors
entering or operating in host countries and would grant them national
treatment. This would involve the removal of performance requirements
and the adoption of a range of investor rights.
However, most developing countries are reluctant to agree to this because
several believe that an investment regime is inappropriate within a trade
organization. Others have concerns about the loss of autonomy over invest-
ment policy and the consequent limitations on industrial policy options.
Also, there are broadly expressed concerns that an investment agreement

would divert time and human resources from other vital work in the WTO.
The fundamental premise of the argument in favor of a multilateral invest-
ment agreement is that it will increase investment flows to developing
countries. An agreement which improves investor protections may stimulate
domestic investment and alleviate the concerns of foreign investors.
However, there are several reasons to be cautious about the responsive-
ness of investment to new multilateral protections. First, the current
absence of multilateral investment disciplines and the failure of previous
attempts to establish them (such as the OECD’s ill-fated Multilateral
Agreement on Investment, MAI) has not deterred foreign investment.
Foreign direct investment has grown rapidly over the last decade, outpacing
both trade and output growth.
Additionally, the absence of a multilateral agreement has not prevented
substantial unilateral liberalization of investment regimes. UNCTAD
reports that between 1991 and 2001, a total of 1,393 changes were made to
national investment regulations. More than 90 per cent of these were liber-
alizing changes. Figure A2.1 shows that in 2001, over 200 regulatory changes
were made in 71 countries, only 6 per cent of which were restrictive changes.
In this environment there does not seem to be a compelling rationale to
force national governments to adopt a uniform multilateral agreement.
Idiosyncratic national regimes are sensitive to national development priorities
and can be tailored to existing institutional arrangements to minimize
implementation costs.
A third reason for caution comes from the historical experience of invest-
ment treaties in generating increased investment flows (see Fig. A2.2).
(a) 80 (b) 250 More favorable to FDI

70 Less favorable to FDI
60 200
50 150
40
30 100
20 50
10
0 0
91
92
93
94
95
96
97
98
99
00
01
91
92
93
94
95
96
97
98
99
00
01
19
19
19
19
19
19
19
19
19
20
20
19
19
19
19
19
19
19
19
19
20
20
Figure A2.1. Liberalization of investment regimes, 1991–2001

(a) Number of countries changing their investment regulations
(b) Number of liberalizing and restrictive changes to investment regulations
(a) 500 North–south (b) 0.3

South–south
400 Transition Europe
0.2
300
200
0.1
100
0 0
1960s 1970s 1980s 1990s 3 2 1 1 2 3
Years before Year signed Years after
signing signing
Figure A2.2. Bilateral investment treaties, 1960s–1990s

(a) Growth of BITs in developing countries (number of treaties in decade)
(b) The effect of BITs on FDI (share of annual FDI flow)
Source: World Bank (2003).
Bilateral investment treaties (BITs) surged in the 1990s to more than 2,000
in 2001. There has been significant activity between developing countries,
which accounted for 42 per cent of new BITs in 2001 (UNCTAD 2002). As
noted earlier, BITs proscribe a range of investment protections that often go
further than many of the realistic proposals before the WTO. Yet there is not
much evidence that the signing of bilateral investment treaties increases
the flow of investment. An UNCTAD (1998) study found no relationship
between the level of FDI and the number of BITs signed by host countries.
A more comprehensive study by Hallward-Driemeier (2002) looked at the
bilateral flows of OECD countries to 31 developing countries over twenty
years. After accounting for trends, they found little evidence that BITs
increased investment to developing countries. More research needs to
be done on the effects of investment treaties on investment volume, but
the existing evidence suggests that the benefits of additional treaties may
be small.
Fourthly, the most serious concern, nationalization of foreign invest-
ments, has already been addressed at both the national and international
level, though national and multilateral agencies providing guarantees
against such confiscations of property (MIGA, the Multilateral Investment
Guarantee Association, is part of the World Bank Group). Going beyond this
entails difficult judgments about what are ‘legitimate’ and illegitimate
restrictions. Most fundamentally, each country has an incentive to develop
an investment regime balancing provisions which might serve to attract
more foreign investment with those protecting against potential adverse
effects on the citizenry. International agreements might help developing
countries provide assurances that they will abide by their commitments

(which is what MIGA has responsibility for), but they should not dictate
how each country should make that balance.
A further difficulty is that in providing further protection, even bilateral
agreements negotiated by trade ministers often go too far, intruding on
national sovereignty in unacceptable ways. The problem is that it often takes
years before the full import is discovered, by which time the possibility of
revising the treaty has become difficult and tendentious. The infamous
Chapter 11 of NAFTA provides a compelling case in point, where foreign
investors were given more rights than domestics (e.g. for compensation for
changes in costs or profitability as a result of even fully justified regulations,
in what are called regulatory takings). This is arguably having an adverse
effect in the development of important regulations in areas like the environ-
ment and consumer protection. (The Clinton Administration was opposing
attempts to provide such compensation in domestic legislation even as
its trade negotiators were putting such a provision into NAFTA, without
seeking prior approval of the Cabinet or the National Economic Council. Once
such provisions are put into an agreement, it is hard to take them out.) Even
judicial protections, such as punitive damages, have come under question.
The attempt to impose restrictions on capital market liberalization
illustrates the dangers of these non-trade-related investment agreements.
(Trade-related capital flows are already covered within current IMF agree-
ments.) There is compelling evidence that full liberalization has little effect
on economic growth, but exposes countries to increased instability—a fact
recognized even by the IMF in its recent Board paper.
For these reasons, the current direction of WTO negotiations on invest-
ment disciplines seems to offer few advantages to developing countries.
An international agreement on investment rules of the type currently being
proposed is ultimately designed to maximize foreign investors’ rights while
minimizing the authority of governments in developing countries. Instead
the WTO should focus on improving the investment environment in ways
which strengthen the bargaining position of governments vis-à-vis foreign
investors rather than weakening it.
One area in which there is clear cause for multilateral action is the reduc-
tion of ‘beggar-my-neighbor’ investment incentive competition for foreign
investment. Since the mid-1980s the efforts of national and subnational
levels of government to attract direct investment to their jurisdictions have
increased considerably (Charlton 2003).
Political pressure on governments to be seen as ‘job winners’ push policy-
makers to play a race-to-the-bottom game. Oxfam (2000) estimates that
developing countries lose US$35 billion per year due to a competitive pres-
sure to reduce corporate tax rates combined with the transfer of profits out
of developing countries to low-tax environments.
The potential negative consequences of investment competition are
particularly acute in developing nations. The risk of ‘overbidding’ is exacer-
bated by institutional weaknesses, poor cost–benefit analysis and, in some
cases, corruption. Moreover, the potential consequences of excessively
generous incentives might be increased in those developing nations where
fiscal positions are already weak.
Agreements to limit the size of incentives seems to be the most obvious
approach to pursue within a multilateral framework. The European Union
provides a good example of the kind of approach to policy coordination that
might benefit developing countries. The EU has been operating state aid
guidelines now for several decades. Although grants to foreign direct invest-
ment are not explicitly targeted by Commission policy, in practice they are
one of the main forms of state aid regulated by it. The EU takes the general
view that state aid is incompatible with the common market. The definition
of state aid clearly encompasses traditional instruments of investment
attraction. Indeed the European Commission classifies state aid as including
(1) grants to firms; (2) loans and guarantees; (3) tax exemptions; and
(4) infrastructure projects benefiting identifiable end-users. The European
Commission claims some success in reducing subsidies in the EU. There is
evidence that the Commission has used its guidelines to effectively restrict
incentives in some areas. For example, before the introduction of guidelines
for the support of small and medium-sized enterprises, it was not rare to find
state-aid grants of as much as 20 per cent of an investment project. Under
the new framework, the fixed maxima are 7.5 per cent for medium-sized
enterprises and 15 per cent for small enterprises.1
A2.2 Competition policy
Strong competition policy backed by clearly enforced laws is beneficial to

developing countries and should be encouraged in international forums.
There is a clear concern that the benefits of a liberal trade regime would
1
An example from the Czech Republic provides an illustration of how the Commission uses its power in
practice. The Czech Republic planned to offer subsidies to the Volkswagen unit Skoda for an engine plant at
Mlada Boleslav. After a year of negotiations with the EU, the government agreed to slash tax breaks and
grants that it was offering to Skoda from US$120m to US$22m.
be undermined by domestic or international monopolies and cartels.

Liberalization might largely simply transfer rents that had been accruing
to the government to private sector monopolies, and not lead to lower
consumer prices. However, competition policy disciplines as envisaged by
the proponents of a WTO agreement may impinge on the ability of each
country to choose a competition policy model which is suitable for its own
development priorities, and the proposals under discussion do not address
the most important concerns of developing countries. What is required is a
paradigm for viewing competition from a development perspective. It is
important to ensure that developing country producers receive treatment
equal to that of domestic producers (which they currently do not), and
to ensure that developing country consumers can be protected from non-
competitive actions by global anti-trust action, including anti-trust actions
centered in the developed countries. National competition law and policy
should complement other national development objectives (such as indus-
trial development). Moreover, such law and policy should not hinder
government efforts to minimize adjustment costs resulting from structural
change generated by WTO-driven liberalization in other areas. Some sensitive
industrial sectors may require protection from advanced foreign firms
for the time it takes to create local capacity. In addition, it needs to be
recognized that the legal frameworks which have been developed in the
advanced industrial countries to promote competition are costly to admin-
ister. Early on, there was an awareness of the risk of politicization of com-
petition policy, providing one of the rationales for private enforcement
actions. There is some reason to believe that those fears have been justified,
and when competition policy is incorporated within a trade regime, there is
often more a concern for the promotion of the interests of the country’s cor-
porations than on the well-being of consumers. Thus, a failure of a country’s
firms to do well in a market will be blamed on anti-competitive actions,
but the attempt by a foreign government to protect its citizens from the
anti-competitive practices of one’s own company will be viewed with
suspicion. Because of their costs, however, private enforcement actions
are often not feasible for those harmed in developing countries.
For these reasons, some of the conventional models of competition
enforcement which operate in developed countries may not be appropriate
for a developing country. In the discussion below, we identify some policies
that might redress the imbalance.
The theoretical benefits of the maintenance of competition are clear.
Indeed, the benefits that are associated with free markets are only enjoyed
if those markets are competitive. As we suggested before, trade liberaliza-
tion and privatization will only deliver on their promised benefits in a
competitive environment. But imperfections in competition are pervasive,

especially in developing countries, where markets are often small. There is
an abiding concern that a large multinational can use its economic power
to become dominant in certain markets in developing countries—these
companies often have global sales that exceed the GDP of the economy.
While the case for strong competition policy is clear, there is regrettably
only a small amount of evidence on the welfare effects of competition policy
itself. Kee and Hoekman (2002) investigate the impact of competition law
on estimated industry mark-ups over cost. They use time series panel data
from 28 industries in 42 countries. They conclude that anti-trust legislation
has no individual impact on the size of mark-ups. By contrast they conclude
that imports and lower entry barriers (which anti-trust policies can lead to)
are associated with a larger payoff, a result supported by several studies
(Djankov, La Porta, Lopez-de-Silanes et al. 2002; Hoekman, Kee, and Olarreaga
2001; Vandenbussche 2000).
By contrast, Clarke and Evenett (2003) show that in Latin America, Asia,
and Western Europe, jurisdictions that did not enforce their cartel laws
suffered greater overcharges than those nations that actively enforced them.
New competition enforcement regimes are associated with significant
implementation costs. Competition law is technical and requires institu-
tional skills and resources that are in short supply in many developing coun-
tries. In addition, competition law enforcement is expensive. OECD and
national sources indicate that the annual budget of the anti-trust office in
OECD countries is in the range of US$15–50 million plus. For developing
countries with enforcement agencies the budgets are lower but still signifi-
cant (Hoekman and Mavroidis 2002).2
Discussions under ‘competition’ center around two issues: preventing
monopolization and anti-competitive practices, and preventing govern-
ments from acting in ways which give an ‘unfair’ competitive advantage
to their firms, either by imposing requirements on foreign firms or by
subsidizing their own firms.
Under the first rubric, preventing monopolization and anti-competitive
practices, there are two important reforms. The first is the adoption of a
single standard for predatory anti-competitive behavior between domestic
and foreign firms. There is a large literature on the cost of dumping laws,
and on their inequities (including those associated with the manner in
which they are implemented).3
2
For example, 20 noted earlier, the costs of anti-trust offices are large in Mexico (US$14m), Poland
(US$4.1m), Argentina (US$1.4m), and Hungary (US$2m).
3
See for example Martinez-Giralt and Barros (1997).
Another priority should be to protect purchasers in developing countries

from paying excessively high prices as a result of monopolies and especially
international cartels. National competition policy may ignore collusion by
domestic firms to raise prices in export markets since there is no harm to
those within the country. Moreover, developing countries without the
resources to enforce competition policy effectively on international firms
may suffer from international cartels. There is a small amount of (mainly
informal) evidence on the effects of international cartels on developing
countries (see Figure A2.3). Levenstein, Oswald, and Suslow (2002) analyse
sixteen goods whose supply was found to be internationally cartelized by
American or European firms. They found that in 1997 developing countries
purchased US$36.4 billion of goods from a set of ten industries that had
seen a price-fixing conspiracy in the 1990s. This amounts to 2.9 per cent of
developing countries’ total imports which may have been subject to collu-
sive price-fixing by firms from developed countries. It has been estimated
that cartels in developed countries cost consumers in developing countries
up to US$7 billion in the 1990s. Some of the worst-offending cartels have
been found in the international maritime transport industry. Such cartels
are often approved by national competition authorities but have been
found to lead to higher prices for consumers. Fink, Mattoo and Rathindran
(2001) estimate that collusive practices in the maritime transport industry
have cost consumers in the US alone up to US$2.1 billion. If developing
countries were to save the same proportion of their shipping costs the sav-
ings would be US$2.3 billion.
35
Imports to developing countries
30 Imports to industrial countries
25
20
15
10
15
0
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000
Figure A2.3. Volume of imports affected by cartels, 1981–2000

Total imports in sixteen products where cartels have been proved to exist
Source: GEP (2003).
In the mid-1990s, as Russia moved from communism to a market

economy, one of the few goods that it could easily sell internationally was
aluminum, but its attempts to enter Western markets were hampered as
American aluminum companies put pressure on the United States to create
an international aluminum cartel that would strictly limit the extent to
which Russia could enter Western markets and which would thereby main-
tain international prices at high levels. This is a case in which both Russia
and consumers around the world were injured. But those affected had no
recourse to anti-trust laws, especially since governments were involved
in the very creation of the cartel.
Almost undoubtedly, the most important international cartel is OPEC,
which attempts to control the price of oil. Though several OPEC members
are developing countries, most of those in the developing world live in
countries which are net importers of oil and therefore are worse off as a
result of the oil cartel.
This suggests there might be potential gains from multilateral action to
ban export cartels, including those in which governments are a party.
A key issue is how developing countries can respond to anti-competitive
actions, including export cartels. No small developing country could force
the break-up of a cartel; each would find it expensive to bring a court
case, and even if one of them could bring such a case in its own jurisdiction,
enforcement of a judgment would be difficult. One option would be to allow
governments of detrimentally affected countries to use the court system
of OECD countries to prosecute offending firms. Further, adversely affected
parties (including governments acting on behalf of citizens) should be
allowed to be a plaintiff in civil actions in the courts of the advanced
industrial countries, and there should be a provision for class-action suits,
with flexible standards for class certification, which would in particular
encourage purchasers of products in foreign countries to be able to join
in with each other and with other purchasers in the advanced industrial
countries. Given the limited resources of developing countries and the
high costs of suits, assistance from OECD countries would be desirable.
Hoekman and Mavroidis (2002) suggest the creation of a ‘special prosecutor’
within the WTO with authority to bring cases in the relevant jurisdiction
on behalf of developing country consumers.
Merger policy is another area in which national competition policies may
have international spillover effects. The concern is that nationalistic
approval criteria may allow mergers between domestic firms even when the
global welfare effect is negative, so long as the welfare benefits within the
country are positive. If these firms have a larger combined market share in
some smaller foreign markets than they do in the domestic market,

a merger may be domestically acceptable but globally undesirable.
Both from the perspective of developed and less developed countries, there
is concern about attempts to harmonize national competition policies, for,
in doing so, there is a real risk that the ‘least common denominator’ will be
accepted, that is, one which will provide the least protection for consumers.
Even a movement towards a common framework which might be close to a
‘least common denominator,’ would risk overlooking the differing circum-
stances of the developing countries and encouraging governments to move
towards this low standard. This is especially the case because corporate
interests tend to be far better represented in trade negotiations than con-
sumer interests.
It would be better to use the discussion of competition issues within the
WTO to encourage countries to develop high competition standards and to
recognize some of the ways that countries may legitimately do so. For instance,
the United States has, on several occasions, recognized that concerns about
competition ‘trump’ intellectual property; standard intellectual property
protections, which give temporary monopoly power, have been circum-
scribed when they lead to excessive monopolization of particular markets.
This is a principle which should be more widely recognized: it should be
made explicit that such actions are not a violation of the TRIPS Agreement.
As a second example, per se rules may be easier and less costly to enforce
than ‘rule of reason’ judgments, which require the careful balancing of costs
and benefits of certain potentially anti-competitive practices. Strong per se
rules (such as limiting the share of the country’s market that any firm may
have) should be allowed, even if they have the effect of limiting entry by
international firms (for whom, say, entry into a market would only be
worthwhile if they had a dominant position).
On the second rubric, actions taken to give a competitive advantage to a
country’s own firms (or impose a competitive disadvantage on foreign
firms), development concerns should be given priority, and actions which
arguably have a development objective should be allowed, at least on a
medium-term basis, even if they put foreign firms at a disadvantage.
Inevitably, firms from developing and developed countries are in different
circumstances—each has some advantages over the others. Firms from
advanced industrial countries often have access to lower-cost capital and to
government-financed research, especially by-products of the huge expendi-
tures on defense. Local firms may have more local knowledge, and that may
give them a competitive advantage. Inevitably, there will be some vagueness
in what is meant by ‘levelling the playing field’. But approaching the issue
from the development perspective provides some guidance into what kinds
of policies should be allowed by developing countries. Developing countries
should be allowed to provide capital to domestic firms at ‘reasonable terms’,
which may be significantly below the very high interest rates that are
imposed on them as a condition of IMF loans, or which they feel they have
to maintain in order to prevent a currency crisis. Imposing ‘community
reinvestment act’ requirements on banks to lend a certain fraction of
their money to particular groups, e.g. underserved minorities, or small or
medium-sized domestic enterprises, is a legitimate restriction. Giving
preferences to small and medium-sized enterprises for government
contracts (see below) should also be legitimate, even if in doing so, multi-
national firms are in effect discriminated against. At the same time, the
standards for judging whether the provision of infrastructure which, in the
first instance, may be directed at a particular enterprise is a subsidy or not
should be looked at from different perspectives for a developing country
than for a developed country. While advanced industrial countries have
long been critical of local content requirements, such requirements may
in fact facilitate developmental objectives.
A2.3 Government procurement
Government purchases of goods and services are a significant proportion

of world GDP. Recent analysis by the OECD indicates that total central
government expenditure of OECD countries was almost US$2 trillion in
1998.4 In developing countries this figure was US$300 billion—equal to six
times the total annual multilateral and bilateral aid currently given to
developing countries (Evenett 2003).
In an attempt to harness this part of the international economy, several
WTO members signed the ‘plurilateral’ (only binding to those members that
choose to sign) Agreement on Government Procurement (GPA) at the
Uruguay Round in 1994. One of the GPA’s primary long-term objectives is
to ensure that government decisions to purchase goods and services do not
depend on the location of production or the affiliation of the supplier.
Many developed countries, principally the US and the EU, would like to
see the GPA develop into a multilateral agreement which in the first stage
draws all members into an agreement on transparency, and in the second
4
This figure excludes military expenditure and the payment of state employees.
stage extends the scope to due process and national treatment for foreign
firms.
Estimates of the value of a broad multilateral procurement agreement
(encompassing both transparency and non-discrimination) depend on the
size of the government procurement market, the size of preference margins
extended to domestic suppliers, and the general equilibrium gains and
losses derived from the elimination of these preferences.
As described above, government procurement accounts for an average
of about 10–15 per cent of GDP for developed countries and as much as
20 per cent of GDP for developing countries. National domestic preference
margins are estimated by analysing either national policies or the price
wedges that explain government purchasing choices. Francois, Palmeter,
and Nelson (1997) estimate the margin of preference for OECD countries to
be in the 13–50 per cent range.
However, the proportion of government purchases whose prices are
inflated by preferential treatment may not be large. In a procurement
auction, preferences only raise prices when domestic bidders are not the
lowest but are within the preference margin, or when the domestic bidders
are the lowest but there is only one of them (in which case the domestic
bidder raises its sell price to the lowest foreign price plus the preference
margin). However, in other circumstances a preferential procurement
policy may actually reduce procurement costs. McAfee and McMillan
(1989) show that preferential polices can cause foreign suppliers to reduce
their sell price so as to bid under the domestic firms receiving preferences.
Government procurement policies have important economic and social
roles in developing countries which would be curtailed if governments
were mandated to observe national treatment principles. The level of
expenditure and the attempt to direct the expenditure to locally produced
materials is a major macroeconomic instrument, especially during reces-
sionary periods, to counter economic downturn (Kohr 2003). If the foreign
share increased, there would be a leakage in government attempts to boost
the economy through increased spending during a downturn.
Additionally, procurement policies might be used to boost domestic
industries or encourage development in specific sectors of national interest.
Social objectives could also be advanced by preferences for specific groups or
communities, especially those that are under-represented in economic
standing.
Finally, while developed countries (and especially, particular firms
within developed countries) have much to gain from increased access to
government procurement in developing countries, it is not likely that the
gains are fully reciprocal. Even the EU has had difficulty making inroads
into US government defense procurement. Furthermore, the consequences
of procurement liberalization will inevitably depend on liberalization of
services and labor flows. If the US government decides to subcontract army
meal services, will it allow a foreign firm to provide the service? Almost
surely, security concerns will demand that the firm be American. But since
defense is a larger portion of expenditures for America, it means that a larger
portion of government procurement will be exempt. Some communities in
the United States have moved towards contracting out a range of public ser-
vices, including schools and prisons. It is unlikely that foreign firms have
equal access.
In the context of the important government objectives achieved by pro-
curement policy and the lack of any externalities to justify international
action, it seems important that developing countries retain their autonomy
over this area of policy.5
A2.4 Trade facilitation
Trade facilitation initiatives hold out the promise of increasing trade and
efficiency by reducing onerous trade-related costs.6 Such costs include:
regulatory compliance costs; charges for trade-related services; procedural
delays; lack of predictability; and lost business opportunities (Lucenti
2003).
The benefits of improving trade facilitation include: increasing trade in
goods and services; promoting competition, which can spur productivity
gains as well as lower prices; enhancing efficiency in both the state
sector and the private sector; improving the business environment and so
encouraging foreign direct investment; and increasing participation of
small and medium-sized enterprises in international trade.
5
Moreover, it will be difficult to implement any procurement policy in a way which would be widely
acceptable. For example, since a large proportion of American expenditures are for defense, it will almost
surely claim a defense exception. Much of government procurement is for services, but the provision of
such services requires the temporary movement of natural persons. Contract specification can often be
used to give local firms a preference; ascertaining whether the contract specification is ‘reasonable’ will be
extremely tendentious.
6
The Doha Declaration (para. 27) states that until the Fifth Ministerial Conference, the WTO Council for
Trade in Goods shall review and as appropriate clarify and improve relevant aspects of Arts. V, VIII, and X of
GATT 1994 and identify the trade facilitation needs and priorities of members, particularly developing
and least developed countries. (Article V is on freedom of transit, Art. VIII concerns fees and formalities
connected with import and export, and Art. X is on publication and administration of trade regulations.)
The empirical evidence below suggests that many of these benefits may
be economically significant but are associated with large implementation
costs. Rather than imposing new obligations within the WTO, progress
on trade facilitation should be achieved through national efforts aided by
technical assistance.
Few studies have explicitly examined the potential gains from trade
facilitation. There are dramatic anecdotal stories: e.g. Costa Rica’s com-
mitment to trade facilitation was arguably critical in attracting the large
Intel plant. Modern manufacturing has increasingly relied on just-in-time
inventory methods, and these cannot operate if there are costly delays at
borders.
Moreover, the studies presented below differ in terms of the scope of trade
facilitation considered and the breadth of countries and commodities
analysed. This makes their results difficult to compare meaningfully.
However, the results below do highlight the magnitude of potential gains
from trade facilitation.
Ernst and Whinney (1987) surveyed EC business costs for the European
Commission. The customs compliance costs associated with intra-EC trade
were estimated to be 1.5 per cent of the total value of trade between mem-
ber countries.
The US National Committee on International Documentation (US NCITD)
analysed the benefits of trade facilitation in a 1971 study, subsequently
updated in the 1990s by Raven (1996). These studies found that the costs of
documentation and compliance with export and import regulations (at both
ends of the transaction) represented more than 7.5 per cent of the total value
of US shipments.
A study by the Swedish Trade Procedures Council (SWEPRO) in 1995
used data from companies and government sources to estimate the cost of
compliance with Swedish trade procedures. It concluded that these costs
could be as much as 4 per cent of the value of imports and exports.
More recently, a study by Wilson, Mann, and Otsuki (2003) used a
computable general equilibrium framework to estimate gains from trade
facilitation. They estimated the effect of bringing the below-average Asia
Pacific Economic Cooperation (APEC) members halfway to the APEC
average in four key areas of trade facilitation (administrative transparency,
e-commerce, logistics, and standards harmonization). They estimate this
type of facilitation would increase APEC trade by US$280 billion.
Wilson et al. compare the relative benefits from trade facilitation with
those from traditional market access initiatives. They estimate that a
reduction of all tariffs to zero from an APEC average of 6.5 per cent would
create a gain of US$27.8 billion. By comparison they find that the improve-
ment in trade facilitation necessary to achieve the same gains is small.
Relatively minor improvements in port efficiency, customs procedures, and
e-business usage deliver similar-sized gains.
This brief survey suggests that there is a wide span of estimates regarding
the costs of trade procedures, ranging from 1.5 to 7.5 per cent of the value of
trade flows.
The implementation costs associated with realizing gains from trade
facilitation are also significant. Administrative changes are associated with
obvious costs to both governments (the creation of new systems and
enforcement of new regulations) and business (compliance). For developing
countries, a large part of the costs to government should be borne by
technical assistance from developing countries.
The costs of trade facilitation depend on the type of reform proposed. For
example, as noted earlier, the World Bank assisted Tunisia in its program of
streamlining and modernizing its customs procedures. The total value of
World Bank loans to Tunisia for this purpose was US$35 million in 1999.
Similarly, the World Bank lent US$38 million to Poland for upgrading the
physical and managerial infrastructure of its port facilities (Wilson 2001 and
WTO 2000). In some cases, streamlining procedures will both facilitate
trade and save the government money.
Developing countries should therefore put forward the view that
improvements in trade facilitation should be made through national efforts
aided by technical assistance, rather than through imposing additional
obligations in the WTO. If the consideration of the problems in these areas
results in some agreements, these should, at best, be adopted only as guid-
ing principles or as flexible best-endeavour provisions, not enforceable
through the dispute settlement process (Das 2002).
Discussion here, as elsewhere, has focused more on the concerns of
developed countries in gaining access to developing countries than on what
should be the concern of the Development Round, i.e. developing countries
gaining access to developed countries’ markets. Of intense concern in the
last few years have been America’s visa restrictions. Such restrictions
may or may not be justified by legitimate concerns for security. However,
it is certainly true that the existence of these security concerns illustrates
that the developed countries often put non-trade concerns over trade con-
cerns but often criticize developing countries when they attempt to do the
same. Yet the inability to send businessmen to America to market their
goods and the long delays and high costs (especially in time) in obtaining a
visa are clearly important impediments to trade. This is an issue that would
be high on a development-oriented agenda for trade facilitation. The fact
that it has not been on the agenda at all is just another illustration of the gap
between what has been sold as a Development Round agenda and an agenda
that would truly promote the interests of the developing world.
References
3
ABARE (2001), ‘Export Subsidies in the Current WTO Agriculture
Negotiations’, ABARE Current Issues 1:6, 1–8.
Anand, Sudhir, and Vijay Joshi (1979), ‘Domestic Distortions, Income
Distribution and the Theory of Optimum Subsidy’, The Economic
Journal 89, 336–52.
Anderson, K. (2003), How Can Agricultural Trade Reform Reduce Poverty?,
Centre for International Economic Studies, University of Adelaide,
Discussion Paper 0323 (Adelaide: Centre for International Economic
Studies).
—— B. Dimaranan, J. Francois, T. Hertel, B. Hoekman, and W. Martin
(2001), ‘The Cost of Rich (and Poor) Country Protection to Developing
Countries’, Journal of African Economies 10, 227–57.
—— and S. Yao (2003), ‘How Can South Asia and Sub-Saharan Africa Gain
from the Next WTO Round?’, Journal of Economic Integration 18,
466–81.
APEC Secretariat (1997), ‘The Impact of Trade Liberalization in APEC’,
paper presented at Asia Pacific Economic Cooperation (APEC) Conference,
1988.
Auboin, M., and M. Meier-Ewert (2003), Improving the Availability of Trade
Finance during Financial Crises, WTO Discussion Papers 2 (Geneva,
World Trade Organization).
Baldwin, Robert E. (1969), ‘The Case against Infant-Industry Tariff Protection’,
Journal of Political Economy, 77, 295–305.
—— John Mutti, and J. David Richardson (1980), ‘Welfare Effects on the
United States of a Significant Multilateral Tariff Reduction’, Journal of
International Economics 10, 405–23.
Beghin, J. C., D. Roland-Holst, and D. van der Mensbrugghe (2003), How
will Agricultural Trade Reforms in High-Income Countries Affect the
Trading Relationships of Developing Countries? (Paris: OECD).
Ben-David, Dan (1993), ‘Equalizing Exchange: Trade Liberalization and
Income Convergence’, Quarterly Journal of Economics 108, 653–79.
280 REFERENCES
Berg, A., and A. O. Krueger (2002), ‘Trade, Growth and Poverty’, paper
presented at the Annual World Bank Conference on Development
Economics, Washington, DC.
Bergsten, C. F. (1998), ‘Fifty Years of the GATT/WTO: Lessons from the Past
for Strategies for the Future’, paper presented to the WTO/Graduate
Institute of International Studies symposium “Fifty Years: Looking Back,
Looking Forward”, Geneva.
Besley, T., B. Burgess, and I. Rasul (2000), ‘How Much should Governments
Spend on Social Safety Nets?’, in Social Protection Anchor (Washington,
DC: World Bank).
Bhagwati, Jagdish (2002a), Free Trade Today (Princeton: Princeton University
Press).
—— (2002b), ‘Intellectual Property Protection and Medicines’, The Financial
Times, Sep. 2002.
Bjornskov, C., and K. M. Lind (2002), ‘Where do Developing Countries Go
after Doha? An Analysis of WTO Positions and Potential Alliances’,
Journal of World Trade 36, 543–62.
Blackhurst, R., Lyakurwa, B., and Oyejide, A. (2000), ‘Options for Improving
Africa’s Participation in the WTO’, World Economy 23, 491–510.
Bonaglia, Federico, and Fukasaku, Kichiro (2002), Trading Competitively:
A Study of Trade Capacity Building in Sub-Saharan Africa (Paris: OECD
Development Centre).
Borrell, B. (1999), ‘Bananas: Straightening Out Bent Ideas on Trade as Aid’,
paper presented at the World Bank/WTO Conference on Agriculture and
the New Trade Agenda from a Development Perspective, Geneva.
Bouguignon, F., and C. Morrisson (2002), ‘Inequality among World Citizens:
1820–1992’, American Economic Review, 92, 727–44.
Brenton, P. (2003), ‘Integrating the Least Developed Countries into the
World Trading System: The Current Impact of EU Preferences under
Everything But Arms’, mimeo, World Bank.
Brown, D. K., A. V. Deardorff, A. K. Fox, and R. M. Stern (1995), ‘Computational
Analysis of Goods and Services Liberalization in the Uruguay Round’, in
W. Martin and L. A. Winters (eds), The Uruguay Round and the
Developing Economies, World Bank Discussion Paper 307 (Washington,
DC: World Bank), 365–80.
—— —— and R. Stern (2002), Computational Analysis of Multilateral
Trade Liberalization in the Uruguay Round and Doha Development
Round, University of Michigan Research Seminar in International
Economics, Discussion Paper 489 (Ann Arbor: University of Michigan
Research Seminar in International Economics).
REFERENCES 281
Busch, Marc L., and Eric Reinhardt (2003), ‘Developing Countries and General
Agreement on Tariffs and Trade World Trade Organization Dispute
Settlement’, Journal of World Trade 37, 719–35.
CAFOD Policy Paper, Catholic Agency for Overseas Development.
Carlin, Wendy, and Colin Mayer (2003), ‘Finance, Investment and growth’,
Journal of Financial Economics 69, 191–226.
CEC (2002), ‘Everything but Arms Proposal: Possible Impacts on the
Agricultural Sector’, mimeo, European Commission.
Chang, Ha-Joon (2002), Kicking Away the Ladder: Development Strategy
in Historical Perspective (London: Anthem Press).
Charlton, Andrew (2003), Incentive Bidding Wars for Mobile Investment:
Economic Consequences and Potential Responses, OECD Development
Center Technical Paper 203 (Paris: OECD).
—— (2005), ‘A Proposal for Special Treatment in Market Access for
Developing Countries in the Doha Round’, in John M. Curtis and Dan
Ciuriak (eds), Trade Policy Research 2005 (Ottawa: Department of
International Trade), forthcoming.
—— and Joseph E. Stiglitz (2004), ‘A Development-Friendly Prioritization of
the Doha Round Agenda’ The World Economy 28, 293–312.
Charnovitz, S. (2004), ‘The WTO’s Problematic “Last Resort” against
Noncompliance’, mimeo, World Trade Law.
Clarke, J. L., and S. J. Evenett (2003), ‘A Multilateral Framework
for Competition Policy?’, in State Secretariat of Economic Affairs
and S. J. Evenett (eds), The Singapore Issues and the World Trading
System: The Road to Cancún and Beyond (Bern: World Trade Institute).
Commander, Simon, Mari Kangasniemi, and L. Alan Winters (2004),
‘The Brain Drain: Curse or Boon?’, in R. Baldwin and L. A. Winters
(eds), Challenges to Globalisation (Chicago: University of Chicago
Press).
Corden, W. M. (1957), ‘Tariffs, Subsidies and the Terms of Trade’, Economica
24, 235–42.
—— (1974), Trade Policy and Economic Welfare (Oxford: Clarendon Press).
Das, Bhagirath Lal (2002), ‘The New WTO Work Programme’, paper
presented at a Third World Network forum, Geneva.
Dasgupta, P., A. Sen, and D. Starret (1973), ‘Notes on the Measurement
of Inequality’, Journal of Economic Theory 6, 180–7.
Dasgupta, P., and J. Stiglitz 1972, ‘On Optimal Taxation and Public
Production,’ Review of Economic Studies, 39(1), pp. 87–103.
—— —— (1974), ‘Benefit–Cost Analysis and Trade Policies’, Journal of
Political Economy 82, 1–33.
282 REFERENCES
Dasgupta, P., and J. Stiglitz (1977), ‘Tariffs vs. Quotas as Revenue-Raising

Devices under Uncertainty’, American Economic Review 67, 975–81.
—— —— (1985) , Learning-by-doing, Market Structure and Industrial and
Trade Policies, CEPR Discussion Papers 80 (London: Centre for Economic
Policy Research).
Deardorff, Alan V. (1994), ‘Economic Effects of Quota and Tariff Reductions’,
in The New GATT: Implications for the United States (Washington, DC:
Brookings Institution), 7–27.
Dee, P., and K. Hanslow (2000), Multilateral Liberalization of Services Trade,
Productivity Commission Staff Research Paper (Canberra: Ausinfo).
Desai, Mihir, Devesh Kapur, and John McHale (2001), ‘The Fiscal Impact
of the Brain Drain: The Fiscal Impact of Emigration from India’, paper
presented at the NBER-NCAER Conference, Neemrana, India.
Diao, Xinshen, Eugenio Diaz-Bonilla, and Sherman Robinson (2003),
How Much does it Hurt? The Impact of Agricultural Trade Policies on
Developing Countries (Washington, DC: International Food Policy
Research Institute).
Diaz-Bonilla, E., M. Thomas, and S. Robinson (2003), ‘Trade, Food Security
and WTO Negotiations: Some Reflections on Boxes and their Content’, in
OECD, Agricultural Trade and Poverty: Making Policy Analysis Count
(Paris: OECD), 59–104.
Dimaranan, B., T. Hertel, and R. Keeney (2003), ‘OECD Domestic Support and
the Developing Countries’, paper for the UNU/WIDER project on the impact
of WTO on low-income countries, GTAP Center, Purdue University.
Dixit, A. K., and V. Norman (1980), The theory of international trade
(Cambridge: Cambridge University Press).
Djankov, Simeon, Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei
Shleifer (2002), ‘The Regulation of Entry’, Quarterly Journal of
Economics 117, 1–37.
Dollar, David (1992), ‘Outward-Oriented Developing Economies Really Do
Grow More Rapidly: Evidence from 95 LDCs, 1976–85’, Economic
Development and Cultural Change 40, 523–44.
Easterly, William, Roumeen Islam, and J. E. Stiglitz (2001), ‘Shaken and
Stirred: Explaining Growth Volatility’, in Annual Bank Conference on
Development Economics 2000 (Washington, DC: World Bank), 191–212.
Ebrill, L., J. Stotsky, and R. Gropp (1999), Revenue Implications of Trade
Liberalization, IMF Occasional Paper 180 (Washington, DC: IMF).
Ellerman, David (2003), Policy Research on Migration and Development,
World Bank Policy Research Working Papers 3117 (Washington, DC:
World Bank).
REFERENCES 283
Emran, M. S., and J. Stiglitz (2004a), ‘On Selective Indirect Tax Reform in
Developing Countries’, Journal of Public Economics 89, 599–623.
—— —— (2004b), Price-Neutral Tax Reform with an Informal Economy,
Public Economics Working Paper 0407010, (St. Louis: WUSTL Working
Paper Archive).
Epstein, S. (1995), GATT: The Uruguay Round Agreement and Developing
Countries (Washington, DC: Foreign Affairs and National Defense
Division).
Ernst & Whinney (1987), ‘The Cost of Non-Europe: Border-Related
Controls and Administrative Formalities’, in Research on the Cost of
Non-Europe: Basic Findings, vol. 1 (Brussels: Commission of the
European Communities), 7–40.
ESCWA (2004), ‘Exploring Potential of South–South Agreements including
Global System of Trade Preferences (GSTP)’, ESCWA report to the
UNCTAD forum on ‘Multilateralism and Regionalism: The New Interface’,
São Paulo.
Estevadeordal, A. (2000), ‘Negotiating market access: The Use of the North
America Free Trade Area Agreement’, Journal of World Trade 34, 141–66.
FAO (2003), FAO Papers on Selected Issues Relating to the WTO
Negotiations on Agriculture (Rome: FAO).
Findlay, C., and G. McGuire (2001), ‘Restrictions on Trade in Services for
APEC Member Economies’, mimeo, APEC.
—— and T. Warren, (eds) (2000), Impediments to Trade in Services:
Measurement and Policy Implications (London and New York: Routledge).
Finger, J. M. (2000), ‘The WTO’s Special Burden on Less Developed
Countries’, Cato Journal 19:3, 425–37.
—— and P. Schuler (2000), ‘Implementation of Uruguay Round
Commitments: The Development Challenge’, The World Economy 23,
511–25.
—— and A. Zlate (2003), ‘WTO Rules that Allow New Trade Restrictions:
The Public Interest is a Bastard Child’, paper prepared for the UN
Millennium Project Task Force on Trade.
Fink, C., A. Matoo, and R. Rathindran, (2001), Liberalizing Basic
Telecommunications: The Asian Experience, World Bank Policy
Research Working Paper 2718 (Washington, DC: World Bank).
Fischer, Stanley (2000), lunch address given at the conference on ‘Promoting
Dialogue: Global Challenges and Global Institutions’, American
University, Washington, DC.
Francois, J. F. (1999), ‘A Gravity Approach to Measuring Services
Protection’, mimeo, Erasmus University.
284 REFERENCES
Francois, J. F. (2001), The Next WTO Round: North–South Stakes in New

Market Access Negotiations (Adelaide: Centre for International
Economic Studies; Rotterdam: Tinbergen Institute).
—— and W. Martin (2003), ‘Formula Approaches for Market Access
Negotiations’, The World Economy 26, 1–28.
—— Bradley McDonald, and Hakan Nordstrom (1995), ‘Assessing the
Uruguay Round’, paper presented at the World Bank Conference ‘The
Uruguay Round and the Developing Economies’, Washington, DC.
—— D. Nelson, and N. D. Palmeter (1997), ‘Public Procurement: A Post-
Uruguay Round Perspective’, in B. Hoekman and P. C. Mavroidis (eds),
Law and Policy in Public Purchasing (Ann Arbor: University of Michigan
Press), 105–24.
—— and A. Strutt (1999), ‘Post-Uruguay Round Tariff Vectors for GTAP
Version 4’, mimeo, Tinbergen Institute, Erasmus University.
—— and I. Wooten (2000), ‘Trade in International Transport Services: The
Role of Competition’, Review of International Economics 9, 49–61.
—— H. van Meijl, and F. van Tongeren (2003), Trade Liberalization and
Developing Countries under the Doha Round, CEPR discussion Paper
4032 (London: Centre for Economic Policy Research).
—— —— —— (2004), ‘Trade Liberalization in the Doha Round’, mimeo,
Erasmus University.
Freeman, R., R. Oostendorp, and M. Rama (2001), ‘Globalization and Wages’,
paper in progress, World Bank.
Gallup, John, Jeffrey Sachs, and Andrew Mellinger (1998), ‘Geography and
Economic Development’, paper presented at the World Bank’s Annual
Bank Conference on Development Economics, Washington, DC.
GATT Secretariat (1993), An Analysis of the Proposed Uruguay Round
Agreement, with Particular Emphasis on Aspects of Interest to
Developing Countries, MTN.TNC/W/W122 (Geneva: World Bank).
Greenwald, B., and J. E. Stiglitz (1993), ‘Financial Market Imperfections and
Business Cycles’, Quarterly Journal of Economics 108, 77–114.
Grynberg, R., and R. M. Joy (2000), ‘The Accession of Vanuatu to the WTO:
Lessons for the Multilateral Trading System’, Journal of World Trade,
34:6, 159–73.
Hallward-Driemeier, M. (2002), ‘Bilateral Investment Treaties: Do They
Increase FDI Flows?’, background paper for Global Economic Prospects
2003: Investing to Unlock Global Opportunities (Washington, DC:
World Bank).
Hamilton, C., and J. Whalley (1984), ‘Efficiency and Distributional
Implications of Global Restrictions on Labour Mobility: Calculations and
Policy Implications’, Journal of Development Economics 14, 61–75.
REFERENCES 285
Handbury, J. (2005), ‘Do Rules of Origin Work? An Empirical Assessment of

the Effectiveness of Rules of Origin in NAFTA’, mimeo, Columbia
University, New York.
Harris, J. R., and M. P. Todaro (1970), ‘Migration, Unemployment and
Development: A Two-Sector Analysis’, American Economic Review 60,
126–42.
Harrison, Glenn W., Thomas F. Rutherford, and David G. Tarr (1995),
‘Quantifying the Uruguay Round’, paper presented at the World Bank
Conference ‘The Uruguay Round and the Developing Economies’,
Washington, DC.
Harrison, W. J., and K. R. Pearson (1996), ‘Computing Solutions for Large
General Equilibrium Models Using GEMPACK’, Computational
Hausmann, Ricardo, and Dani Rodrik (2002), ‘Economic Development as
Self-Discovery’, NBER Working Papers 8952 (Washington, DC: National
Bureau of Economic Research).
Hensen, H. T., S. Robinson, and F. Tarp (2002), General Equilibrium
Measures of Agricultural Policy Bias in Fifteen Developing Countries,
TMD Discussion Paper 105 (Washington, DC: International Food Policy
Research Institute).
Hertel, T. (1997), Global Trade Analysis: Modelling and Applications
(Cambridge: Cambridge University Press).
—— (1999), ‘Potential Gains from Reducing Trade Barriers in
Manufacturing, Services and Agriculture’, paper presented at the 24th
Annual Economic Policy Conference, Federal Reserve Bank of St. Louis.
—— and W. Martin (2000), ‘Liberalizing Agriculture and Manufactures in a
Millennium Round: Implications for Developing Countries’, World
Economy 23, 455–69.
—— K. Anderson, J. Francois, and W. Martin (2000), Agriculture and Non-
Agricultural Liberalization in the Millennium Round, Centre for
International Economic Studies, University of Adelaide Discussion Paper
0016 (Adelaide: Centre for International Economic Studies).
—— M. Ivanic, P. Preckel, and J. Cranfield (2003a), The Earnings Effects of
Multilateral Trade Liberalization: Implications for Poverty in Developing
Countries, GTAP Working Paper 16, version 2 (West Lafayette, Ind.: GTAP).
—— —— —— —— (2003b), ‘Trade Liberalization and the Structure of
Poverty in Developing Countries’, paper prepared for the Conference on
Globalization, Agricultural Development and Rural Livelihoods, Cornell
University, Ithaca, NY.
Hoekman, Bernard (1996), ‘Assessing the General Agreement on Trade in
Services’, in W. Martin and L. A. Winters (eds), The Uruguay Round and
286 REFERENCES
the Developing Countries (Cambridge and New York: Cambridge

University Press), 137–68.
Hoekman, Bernard (2003), More Favorable Treatment of Developing
Countries: Toward a New Grand Bargain (Washington DC: World Bank).
—— (2004a), ‘Making the WTO More Supportive of Development’, mimeo,
World Bank.
—— (2004b), Operationalizing the Concept of Policy Space in the WTO,
World Bank Policy Brief 4 (Washington, DC: World Bank).
—— H. Kee, and M. Olarreaga (2001), Markups, Entry Regulation and Trade,
World Bank Policy Research Paper 2662 (Washington, DC: World Bank).
—— and Michael M. Kostecki (1997), The Political Economy of the World
Trading System: From GATT to WTO (Oxford: Oxford University Press).
—— and Petros C. Mavroidis (1994), ‘Competition, Competition Policy and
the GATT’, World Economy 17, 121–50.
—— —— (2002), Economic Development, Competition Policy and the
WTO, World Bank Policy Research Working Paper 2917 (Washington,
DC: World Bank).
—— C. Michalopoulos, and L. A. Winters (2003), ‘Special and Differential
Treatment for Developing Countries: Towards a New Approach in the
WTO’, mimeo, World Bank.
—— F. Ng, and M. Olarreaga (2002), ‘Reducing Agricultural Tariffs versus
Domestic Support: What’s More Important for Developing Countries?’,
mimeo, World Bank.
Home Office (2001), Migration: An Economic and Social Analysis,
Research, Development and Statistics Directorate Occasional Paper 67
(London: Stationary Office).
Horn, H., and P. Mavroidis (2003), ‘Which WTO Provisions Are Invoked By
and Against Developing Countries, CEPR Discussion Paper (London:
Centre for Economic Policy Research).
Hudec, R. E. (1987), Developing Countries in the GATT Legal System,
Thames Essays (London: Trade Policy Research Centre).
Ianchovichina, E., A. Mattoo, and M. Olarreaga (2001), Unrestricted Market
Access for Sub-Saharan Africa: How Much is it Worth and Who Pays?, CEPR
Discussion Paper 2820 (London: Centre for Economic Policy Research).
ILO (2002), Global Employment Trends (Geneva: International Labour
Office).
IMF (1997), World Economic Outlook (Washington, DC: IMF).
—— (2002), Direction of Trade Statistics Quarterly, March (Washington,
DC: IMF).
REFERENCES 287
Isaac, G. E., and W. A. Kerr (2003), ‘Genetically Modified Organisms and

Trade Rules: Identifying Important Challenges for the WTO’, The World
Economy 26, 29–43.
Johnson, H. (1967), Economic Policies toward Less Developed Countries
(Washington, DC: Brookings Institution).
Kaul, Inge, Pedro Conceicao, Katell Le Goulven, and Ronald U. Mendoza
(eds) (2003), Providing Global Public Goods (Oxford: Oxford University
Press).
Keck, A., and P. Low (2004), Special and Differential Treatment in the
WTO: Why, When and How?, Staff Working Paper ERSD-2004–03
(Geneva: Economic Research and Statistics Division, World Trade
Organization).
Kee, Hiau Looi, and Bernard Hoekman (2002), ‘Competition Law and
Market Discipline: A Cross-Country, Cross-Industry Analysis’, mimeo,
World Bank.
Keen, M., and J. E. Lightart (1999), Coordinating Tariff Reduction and
Domestic Tax Reform, IMF Working Paper, WP/99/93 (Washington, DC:
IMF/World Bank).
Kennett, Maxine, Simon J. Evenett, and Jonathan Gage (2005), ‘Evaluating
WTO Accessions: Legal Economic Perspectives’, draft ms.
Ketkar, Suhas, and Dilip Ratha (2000), ‘Development Financing during a
Crisis: Securitization of Future Receivables’, mimeo, Economic Policy
and Prospects Group, World Bank.
Kletzer, Lori G. (2001), Job Loss from Imports: Measuring the Costs
(Washington, DC: Institute for International Economics).
Krueger, Anne (1997), ‘Trade Policy and Economic Development: How We
Learn’, American Economic Review, 87, 1–22.
Kuznets, S. (1955), ‘Economic Growth and Income Inequality’, American
Economic Review 45, 1–28.
Lacarte-Muro, Julio, and Petina Gappah (2000), ‘Developing Countries and
the WTO Legal and Dispute Settlement System: A View from the Bench’,
Journal of International Economic Law 3, 395–401.
Laird, S. (2002), ‘Market Access and the WTO: An Overview’, in Bernard
Hoekman, Aadyita Mattoo and Philip English (eds), Development, Trade,
and the WTO: A Handbook (Washington, DC: World Bank), 97–104.
—— R. Safadi, and A. Turrini (2002), ‘The WTO and Development’, paper
prepared for the Conference on Policy Reform, Tulane University.
Lall, S. (1992), ‘Technological Capabilities and Industrialization’, World
Development 20, 165–86.
288 REFERENCES
Lederman, D., A. Menendez, G. Perry, and J. E. Stiglitz (2000), ‘Mexico—

Five Years After the Crisis’, in Annual Bank Conference on Development
Economics 2000 (Washington, DC: World Bank), 263–82.
Levenstein, Margaret C., and Valerie Y. Suslow (2001), ‘Private
International Cartels and their Effect on Developing Countries’, mimeo,
University of Massachusetts.
—— —— and Lynda J. Oswald (2002), ‘International Price-Fixing Cartels
and Developing Countries: A Discussion of Effects and Policy Remedies’,
mimeo, University of Michigan Business School.
Levine, R. (1997), ‘Financial Development and Economic Growth: Views
and Agenda’, Journal of Economic Literature 35, 688–726.
Levinsohn, J., S. Berry, and J. Friedman (2003), ‘Impacts of the Indonesian
Economic Crisis: Price Changes and the Poor’, in Michael Dooley and
Jeffrey Frankel (eds), Managing Currency Crises in Emerging Markets
(Chicago: University of Chicago Press).
Lewis, W. A. (1955), The Theory of Economic Growth (London, George
Allen & Unwin; Homewood, Ill.: Irwin).
Love, James (2001), ‘Compulsory Licensing: Models for State Practice in
Developing Countries, Access to Medicine and Compliance with the
WTO TRIPS Accord’, paper prepared for the United Nations
Development Programme.
Lucenti, Krista (2003), ‘Is There a Case for More Multilateral Rules on Trade
Facilitation?’, in State Secretariat of Economic Affairs and S. J. Evenett
(eds), The Singapore Issues and The World Trading System: The Road to
Cancún and Beyond (Bern: World Trade Institute), 254–90.
Malcolm, G. (1998), Adjusting Tax Rates in the GTAP Data Base, GTAP
Technical Paper 12 (Lafayette, Ind.: Center for Global Trade Analysis,
Purdue University).
Markusen, J., T. F. Rutherford, and D. Tarr (1999), ‘Foreign Direct Investment
in Services and the Domestic Market for expertise’, paper presented at the
Second Annual Conference on Global Economic Analysis, Copenhagen.
Martin, Philip, and Thomas Straubhaar (2001), ‘Best Practices to Foster
Economic Growth’, mimeo, Cooperative Efforts to Manage Emigration
(CEME), University of California at Davis.
Martin, W. (2001), ‘Trade Policies, Developing Countries, and
Globalization’, in Globalization: Policy Research Report (Washington,
DC: World Bank).
—— and D. Mitra (2001), ‘Productivity Growth in Agriculture and
Manufacturing’, Economic Development and Cultural Change 49, 403–23.
REFERENCES 289
—— and L. A. Winters (eds) (1996), The Uruguay Round and the Developing
Countries (Cambridge and New York: Cambridge University Press).
Martinez-Giralt, Xavier, and Pedro Barros (1997), On the Effects of Anti-
Dumping Legislation, CEPR Discussion Paper 1590 (London: Centre for
Economic Policy Research).
Maskus, K. (2003), ‘Observations on the Development Potential of
Geographical Indications’, background paper prepared for the second
meeting of the Task Force on Trade and Finance of the UN Millennium
Development Goals Project, New Haven, conn.
Massey, Douglas S., Joaquin Arango, Graeme Hugo, Ali Kouaouci, Adela
Pellegrino, and J. Edward Taylor (1998), Worlds in Motion:
Understanding International Migration at the End of the Millennium
(Oxford: Clarendon Press).
Mattoo, A. (2001), ‘MFN and the GATS’, in Trade Policy for Developing
Countries in a Global Economy: A Handbook Washington, DC:
Development Research Group and World Bank Institute, World Bank).
—— R. Rathindran, and A. Subramanian (2001), Measuring Services Trade
Liberalization and its Impacts on Economic Growth: An Illustration,
World Bank Policy Research Working Paper 2655 (Washington, DC:
World Bank).
McAfee, R. P., and J. McMillan (1989), Incentives in Government
Contracting (Toronto: University of Toronto Press).
McDougall, R. (1993), Incorporating International Capital Mobility into
SALTER, SALTER Working Paper 21 (Canberra: Industry Commission).
Melo, Jaime de, and David Tarr (1990), ‘Welfare Costs of US Quotas in
Textiles, Steel and Autos’, Review of Economics and Statistics 72, 489–97.
Michalopoulos, C. (2000), ‘The Role of Special and Differential Treatment
for Developing Countries in GATT and the World Trade Organisation’,
mimeo, World Bank.
Minot, Nicholas, and Francesco Goletti (2000), Rice Market Liberalization
and Poverty in Viet Nam IFPRI Research Report 114 (Washington, DC:
International Food Policy Research Institute).
Newbery, David M. G., and Stiglitz, Joseph E. (1984), ‘Pareto-Inferior Trade’,
Review of Economic Studies 51, 1–12.
Nguyen, T., C. Perroni, and R. Wigle (1993), ‘An Evaluation of the Draft
Final Act of the Uruguay Round’, Economic Journal 103, 1540–9.
Noland, M., and H. Pack (2003), Industrial Policy in an Era of
Globalization: Lessons from Asia (Washington, DC: Institute for
International Economics).
290 REFERENCES
OECD (1993), Assessing the Effects of the Uruguay Round, Trade Policy
Issues 2 (Paris: OECD).
—— (1998), Harmful Tax Competition: An Emerging Global Issue, DAFFE
Report (Paris: OECD).
—— (2002a), Trading Competitively: A Study of Trade Capacity Building in
Sub-Saharan Africa (Paris: OECD Development Centre).
—— (2002b), Agricultural Policies in OECD Countries: Monitoring and
Evaluation (Paris: OECD).
—— (2003), Agricultural Trade and Poverty: Making Policy Analysis Count
(Paris: OECD).
—— (2004), Analysis of CAP Reform, AGR/CA/APM(2003)16/FINAL
(Paris: OECD).
Oman, Charles (2000), Policy Competition for Foreign Direct Investment:
A Study of Competition among Governments to Attract FDI (Paris:
Development Centre Studies, OECD).
Oxfam (2000a), Submission to the Parliamentary Inquiry into Australia’s
Relationship with the World Trade Organisation (Canberra: Oxfam
Australia).
—— (2000b), Tax Havens: Releasing the Hidden Billions for Poverty
Eradication (Oxford: Oxfam).
—— (2003a), Cambodia’s Accession to the WTO: How the Law of the
Jungle is Applied to One of the World’s Poorest Countries, Oxfam
International Briefing Paper (Oxford: Oxfam).
—— (2003b), The EU after Cancún: A Way forward, Oxfam Great Britain
(Oxford: Oxfam).
—— (2004), ‘Extortion at the Gate: Will Viet Nam Join the WTO on
Pro-Development Terms?’, Oxfam International Briefing Paper (Oxford:
Oxfam).
Page, S., and P. Kleen (2004), ‘Special and Differential Treatment of
Developing Countries in the World Trade Organization’, Overseas
Development Institute paper prepared for the Swedish Ministry of
Foreign Affairs.
Pauwelyn, Joost (2000), ‘Enforcement and Countermeasures in the WTO:
Rules are Rules—Toward a More Collective Approach’, American Journal
of International Law 94, 335–47.
Prasad, Eswar, Kenneth Rogoff, Shang-Jin Wei, and M. Ayhan Kose (2003),
‘Effects of Financial Globalization on Developing Countries: Some
Empirical Evidence’, mimeo, IMF.
Prowse, S. (2002), ‘The Role of International and National Agencies in
Trade-Related Capacity Building’, The World Economy, 25, 1235–61.
REFERENCES 291
Rae, Allan N., and Anna Strutt (2003), ‘Doha Proposals for Domestic
Support: Assessing the Priorities’, paper presented at the International
Conference on Agricultural policy reform and the WTO, Capri, Italy.
Raghavan, C. (1999), ‘New Round with an Open-Ended Agenda Advocated’,
INFO-SERVICE, Third World Network, Mar. 1999; available online at
⬍www.twnside.org. sg⬎.
Rama, M. (2003), Globalization and Workers in Developing Countries,
World Bank Policy Research Working Paper 2958 (Washington, DC:
World Bank).
Ranis, G. (1984), ‘Typology in Development Theory: Retrospect and
Prospects’, in M. Syrquin, L. Taylor and L. E. Westphal (eds), Economic
Structure and Performance (Orlando, Fla.; Academic Press), 29–37.
Raven, John (1996), International Trade Procedures: Characteristics and
Costing (London: UK Simpler Trade Procedures Board).
Rodriguez, Francisco, and Dani Rodrik (1999), Trade Policy and Economic
Growth: A Skeptic’s Guide to Cross-National Evidence, National Bureau
of Economic Research Working Paper WP/99/7081 (Cambridge Mass.:
National Bureau of Economic Research).
Rodrik, Dani (1994), ‘Developing Countries often the Uruguay Round’,
paper prepared for the Group d 24, mimeo.
—— (1996), ‘Coordination Failures and Government Policy: A Model with
Applications to East Asia and Eastern Europe’, Journal of International
—— (2001), ‘The Global Governance of Trade as if Development Really
Mattered’, specially invited paper read to the UNDP, New York.
—— (2004), ‘Industrial Policy for the 21st Century’, mimeo, Harvard
University.
Rosenstein-Rodan, P. (1961), ‘Notes on the theory of the “Big push” ’ in
H. S. Ellis and H. C. Wallich (eds), Economic Development for Latin
America (New York: St. Martin’s Press), 202–11.
Rothschild, M., and J. E. Stiglitz (1973), ‘Some Further Results on the
Measurement of Inequality’, Journal of Economic Theory 6, 188–204.
Sachs, Jeffrey D., and Andrew Warner (1995), ‘Economic Reform and the
Process of Global Integration’, Brookings Papers on Economic Activity 1,
1–118.
Safadi, R., and S. Laird (1996), ‘The Uruguay Round Agreements: Impact on
Developing Countries’, World Development 24, 1223–42.
Samuelson, Paul A. (1962), ‘The Gains from International Trade Once
Again’, Economic Journal 72, 820–9.
Sen, A. (1999), Development as Freedom (Oxford: Oxford University Press).
292 REFERENCES
Shapiro, Carl, and J. E. Stiglitz (1984), ‘Equilibrium Unemployment as a

Worker Discipline Device’, American Economic Review 74, 433–44.
Sharer, Robert, Nur Calika, Thomas Dorsey, Paul Ross, Clinton Shielis, and
Piritta Sorsa (1998), Trade Liberalization in IMF-Supported Programs,
IMF World Economic and Financial Surveys (Washington, DC: IMF).
Sharma, R., P. Konandreas, and J. Greenfield, (1996), ‘An Overview of
Assessments of the Impact of the Uruguay Round on Agricultural Prices
and Incomes’, Food Policy 21, 345–50.
South Centre (1996), Liberalization and Globalization: Drawing
Conclusions for Development (Geneva: South Centre).
Srinivasan, T. N. (1998), Developing Countries and the Multilateral
Trading System, Boulder, Colo. and London: Westview Press).
—— and Jagdish Bhagwati, (1999), ‘Outward Orientation and Development:
Are the Revisionists Right?’, Yale University Economic Growth Center
Discussion Paper No. 806.
Stephens, Malcolm (1998), Export Credit Agencies, Trade Finance, and
South East Asia, IMF Working Paper WP/98/175 (Washington, DC: IMF).
Stevens, C. (2002), The Future of Special and Differential Treatment (SDT)
for Developing Countries in the WTO, Institute of Development Studies
(IDS) Working Paper 163 (Brighton: IDS).
Stiglitz, J. E. (1987), ‘On the Microeconomics of Technical Progress,’ in Jorge
M. Katz (ed.), Technology Generation in Latin American Manufacturing
Industries (Basingstoke: Macmillan), 56–77.
—— (1994), ‘Rethinking the Economic Role of the State: Publicly Provided
Private Goods/Replanteamiento del papel económico del estado: bienes
privados suministrados públicamente’, Celección els Llibres Dels Fulls
Econòmics 10, 19–47.
—— (1995), ‘The Theory of International Public Goods and the Architecture
of International Organizations’, Background Paper 7, Third Meeting,
High Level Group on Development Strategy and Management of the
Market Economy, UNU/WIDER, Helsinki.
—— (1996), ‘Some Lessons from the East Asian Miracle’, World Bank
Research Observer 11:2 (Aug.), 151–77.
—— (1997), ‘Dumping on Free Trade: The U.S. Import Trade Laws’,
Southern Economic Journal 64, 402–24, repr. in Sir Hans Singer,
Neelambar Hatti and Rameshwar Tandon (eds), Trips, the Uruguay
Round and Third World Interests (New Delhi: BR Publishing, 1999),
711–48.
—— (1999a), ‘Knowledge as a Global Public Good’, in Inge Kaul, Isabelle
Grunberg, and Marc A. Stern (eds.), Global Public Goods: International
REFERENCES 293
Cooperation in the 21st Century (New York: Oxford University Press),

308–25.
—— (1999b), ‘Trade and the Developing World: A New Agenda’, Current
History 98, 387–93.
—— (1999c), ‘The WTO Millennium Round’, Social Development Review
3:4, (Dec.), 6–9.
—— (2000), ‘Capital Market Liberalization, Economic Growth, and
Instability’, World Development 28, 1075–86.
—— (2001), ‘From Miracle to Crisis to Recovery: Lessons from Four Decades
of East Asian Experience’, in Joseph Stiglitz and Shahid Yusof, Rethinking
the East Asian Miracle (Washington, DC: World Bank), 509–25.
—— (2002), ‘Capital Market Liberalization and Exchange Rate Regimes:
Risk without Reward’, Annals of the American Academy of Political and
Social Science 579, 219–48.
—— (2003), ‘Development-Oriented Tax Policy’, paper presented to the
59th Congress of the IIPF, Prague.
Stolper, Wolfgang, and Paul A. Samuelson (1941), ‘Protection and Real
Wages’, Review of Economic Studies 9, 58–73.
Sutherland, P. (2005), The Future of the WTO: Addressing Institutional
Challenges in the New Millennium, Report by the WTO Consultative
Board to the Director-General (Geneva: WTO).
SWEPRO (1985), Data Interchange in International Trade: Uniform Rules
for Paperless Open Interchange of Trade Information between Different
Computer Systems (Göteborg: SWEPRO).
Tamms, V. (2000), ‘Frontier Cost Estimates of the Impact of Restrictions on
Trade in Air Transport Services’, in C. Findlay and T. Warren (eds),
Impediments to Trade in Services: Measurement and Policy Implications
(London and New York: Routledge).
TWN (2003), ‘US-EC Agriculture Paper Criticised by Developing
Countries’, Third World Network Info Service on WTO Issues, available
online at ⬍www.twnside.org.sg⬎.
Tyers, R., and K. Anderson (1992), Disarray in World Food Markets: A
Quantitative Assessment (Cambridge and New York: Cambridge
University Press).
UNCTAD (1996a), Compendium of the Work and Analysis Conducted by
UNCTAD Working Groups and Sessional Committees on GSP Rules of
Origin, Part I, UNCTAD/ITD/GSP/31 (New York: United Nations).
—— (1996b), ‘Strengthening the Participation of Developing Countries in
World Trade and the Multilateral Trading System’, paper presented at the
ESCAP/UNCTAD/UNDP Meeting of Senior Officials, Jakarta.
294 REFERENCES
UNCTAD (1998), World Investment Report (New York: United Nations).

—— (2000), World Investment Report 2000: Cross-Border Mergers and
Acquisitions and Development, Geneva: United Nations Conference on
Trade and Development.
—— (2001), Duty and Quota-Free Market Access for LDCs: An Analysis of
Quad Initiatives (Geneva: UNCTAD: London: Commonwealth Secretariat).
—— (2002), World Investment Report 2002: Transnational Corporations
and Export Competitiveness, UN publication E.02.II.D.4 (New York and
Geneva: United Nations Conference on Trade and Development).
—— (2003), ‘Main Recent Initiatives in Favour of Least Developed
Countries in the Area of Preferential Market Access: Preliminary Impact
Assessment’ I note by the UNCTAD secretariat for the Trade and
Development Board, fiftieth session, Geneva.
—— (2004), The Least Developed Countries Report 2004, (Geneva: United
Nations Conference on Trade and Development).
UNDP (1997), Human Development Report 1997 (New York and London:
Oxford University Press).
UNDP (2003a), Making Global Trade Work for People (New York: United
Nations Development Program).
—— (2003b), Worker Remittance as an Instrument for Development
(San Salvador: UNDP El Salvador).
UNIDO (2002), Enabling Developing Countries to Participate in
International Trade: Strengthening the Supply Capacity, UNIDO Strategy
Document.
Vandenbussche, Hylke (2000), ‘Trade Policy versus Competition Policy:
Substitutes or Complements?,’ De Economist 148, 625–42.
Verikios, G., and K. Hanslow (1999), ‘Modelling the Effects of Implementing
the Uruguay Round: A Comparison Using the GTAP Model under
Alternative Treatments of International Capital Mobility’, paper pre-
sented at Second Annual Conference on, Global Economic Analysis,
Copenhagen.
—— and Xiao-guang Zhang (2001), ‘The Economic Effects of Removing
Barriers to Trade in Telecommunications and Financial Services’, paper
presented at the Fourth Annual Conference on Global Economic
Analysis, West Lafayette, Ind.
Wainio, J., and P. Gibson (2003), ‘The Significance of Nonreciprocal Trade
Preferences for Developing Countries’, paper presented at the International
Conference on Agricultural Policy Reform and the WTO, Capri (Italy).
Walmsley, T., and L. A. Winters (2002), ‘Relaxing Restrictions on Temporary
Movement of Natural Persons: A Simulation Analysis’, mimeo, University
of Sussex.
REFERENCES 295
Warner, Andrew M. (2002), ‘Once More into the Breach: Economic Growth
and Global Integration’, mimeo, National Bureau of Economic Research.
Warren, T. (2000), ‘The Impact on Output of Impediments to Trade and
Investment in Telecommunications Services’, in C. Findlay and T. Warren
(eds), Impediments to Trade in Services: Measurement and Policy
Implications (London and New York: Routledge), 1–17.
Wilson, John S. (2001), ‘Trade Facilitation Lending by the World Bank—
Recent Experiences, Research and Capacity Building Initiatives’, draft
paper prepared for the WTO Workshop on Technical Assistance and
Capacity Building in Trade Facilitation, Geneva.
—— (2002), ‘Standards, Trade and Development: What is Known and What
do we Need to Know?’ paper presented at the Roundtable on Informing
the Doha Process: New Research for Developing Countries, Cairo.
Wilson, J., C. Mann, and T. Otsuki (2003), Trade Facilitation and Economic
Development: Measuring the Impact, World Bank Working Paper 2988
(Washington, DC: World Bank).
—— —— Yuen Pau Woo, Nizar Assanie, and Inbom Choi (2002), Trade
Facilitation: A Development Perspective in the Asia Pacific Region
APEC.
Winters, A. (2003), ‘Trade Policy as Development Policy’, in J. Toye (ed.),
Trade and Development: Directions for the Twenty-first Century
(Cheltenham: Edward Elgar).
Winters, L. A. (2000), ‘Trade, Trade Policy and Poverty: What are the Links?’,
background paper for the World Bank’s World Development Report
2000/01.
—— (2002), ‘Trade Liberalization and Poverty: What are the Links?’, The
World Economy 25, 1339–68.
—— and Wendy E. Takacs (1991), ‘Labour Adjustment Costs and British
Footwear Protection’, Oxford Economic Papers 43, 479–501.
—— T. Walmsley, Z. Wang, and R. Gyrnberg (2002), ‘Negotiating the
Liberalization of the Temporary Movement of Natural Persons’, mimeo,
University of Sussex.
World Bank (1993a), The East Asian Miracle (Oxford: Oxford University
Press).
—— (1993b), Trade Liberalization: Global Economic Implications
(Washington, DC: World Bank).
—— (2000), Does More International Trade Openness Worsen Inequality?,
World Bank Briefing Paper (Washington, DC: World Bank).
—— (2002a), Global Economic Prospects and the Developing Countries:
Making World Trade for the World’s Poor (Washington, DC: World
Bank).
296 REFERENCES
World Bank (2002b), Trade, Investment, and Development in the Middle

East and North Africa, World Bank report (Washington, DC: World Bank).
—— (2003), Global Economic Prospects, World Bank, Washington, DC.
—— (2004), Global Economic Perspectives 2004: Realizing the Development
Promise of the Doha Agenda (Washington, DC: World Bank).
WTO (2001a), Doha WTO Ministerial 2001: Ministerial Declaration,
WT/MIN(01)/DEC/1 (Geneva: World Trade Organisation).
WTO (2001b), Market Access: Unfinished Business. Post-Uruguay Round
Inventory and Issues, Special Study 6 (Geneva: World Trade Organization).
—— (2003), Adjusting to trade liberalization: the role of policy, institutions
and WTO disciplines, WTO Special Studies 7 (Geneva: World Trade
Organisation).
Zedillo, E. (2001), Report of the High-Level Panel on Financing for
Development, report presented to the UN Monterrey Conference on
Financing for Development (New York: United Nations).
Index
3
Figures, notes and tables are indicated in bold e.g. f, n, t. More than one figure
or table per page is indicated by (a) or (b)
accession negotiations (WTO) 157, agriculture 4, 18, 23, 44, 45–46,

157n, 158, 158f, 159–163 47, 49, 50, 50n, 58, 58n, 59, 62,
accounting systems 181–182 64, 65, 71, 77n, 89–90, 92, 94,
administration costs 182, 191, 192, 101, 107, 109–110, 117,
233, 268 120–124, 160–161, 173,
adjustment assistance 233 178–179, 183, 183f, 184,
adjustment costs 71–72, 77, 90, 91, 199–200, 202, 210, 217–218,
94, 99, 102, 164, 170, 172, 218f, 219–234, 242, 243, 244,
173–195, 188, 203, 205, 213, 246, 257, 260
258, 268 imports, to USA 185t
liberalization 171–214 exports 252, 253f(a), 253f(b)
adjustment programs 177, 203, 208 liberalization 57, 59, 120, 122,
affirmative action programs 155, 240 201–202, 216, 224–225,
Africa 11, 63, 98, 102, 109, 180, 227t, 228t, 229, 235, 245t,
188–189, 195, 210 259t
AIDS pandemic 62 protection 49, 120, 217, 218, 219t,
apparel exports 61 252
drugs 143 elimination 233
legal disputes 84 reforms 58, 120–122, 123, 233
African Growth and Opportunity Act OECD 234n
(AGOA) (USA) 60, 61, 176 subsidies 24, 48, 48n, 69, 76, 80,
Agreement on Agriculture 49, 162 109, 110, 120, 120n, 121n, 162,
Agreement on Customs Valuation 169, 171–172, 223
(WTO) 193 tariffs 219, 220t
Agreement on Government binding 161, 161f, 162; reduction
Procurement (GPA) 273–274 257
Agreement on Safeguards 128 agro-business 224
Agreement on Subsidies 103 allocative efficiency 222, 227, 231,
Agreement on Subsidies and 232, 242, 255, 258
Countervailing Measures America see USA
(SCM) 134, 138–139 Anand, Sudhir 28, 29
Agreement on Textiles and Clothing Anderson, K. 107n, 117n, 121n, 126,
(ATC) 47 218f, 219, 224, 225, 226t, 227,
Agreement on the Sanitary and 228, 229, 229n, 231, 244, 245,
Phyto-sanitary Standards (SPS) 256, 259
162, 192, 211–212 anti-civil strife 135–138
298 INDEX
anti-competitive effects 118–119, Barros, Pedro 269n

144, 147, 148; behind the border rules 103
see also competition Beghin, J. C. 224
anti-competitive practices 269, 272 Ben-David, Dan 34
anti-corruption 134–135; Benin 62
see also corruption Bergsten, C. F. 43, 44n
anti-dumping claims 127; Berry, S. 198
see also dumping duties Besley, T. 204
anti-globalization 36, 39, 54; ‘best information available’ (BIA)
see also globalization 127–128, 130–131
anti-trust legislation 147, 263, 269 Bhagwati, Jagdish 32, 33, 73n, 146
actions 268 bio-piracy 51, 83, 140, 145
enforcement 85, 119, 193n bilateral investment treaties
framework 119, 147 (BITs) 150, 159, 265, 265f
laws (USA) 79n, 130, 147 bilateral trade agreements 5,
Appellate Body (WTO) 128, 136, 164–165, 169–170, 266
136n, 137 Blackhurst, R. 168n
Arango, Joaquin 248 Blandford, David 225
Argentina 19, 109, 122, 123, 139, Blue Box support 58, 59
197, 220, 231, 233 blue tariffs 153;
arms sales 135–136 see also green tariffs; tariffs
Article 2 73 Bolivia 193
Article XVIII 42–43 Bonaglia, Federico 211
Article X1X 128 border protection 102, 123, 230, 233
Article XX 136, 136n Borrell, B. 225, 226
Article 44 91 Botswana 225
Asia 11, 137, 180, 269 brain drain 248
Asia/Pacific countries 189 Brazil 19, 43, 63, 97, 109, 122, 123,
Asia Pacific Economic Cooperation 129, 169, 202, 220, 228, 231,
(APEC) 276–277 233
Asian Development Bank 138 Brenton, P. 61, 61n, 177, 178, 178n,
Atlantic partnership 44 181, 210
Auboin, M. 138 Britain see United Kingdom
Australia 12, 218, 220, 246 Brown, D. K. 52n, 117n, 243, 244,
autarky 12, 25, 26; 255, 256, 257, 258
see also free trade budgets 192, 193
costs 76
Bahamas 188 Burgess, B. 204
balance-of-payments 139 Burkina Faso 62
Baldwin, Robert E. 21, 98n, 174 Busch, Marc L. 83, 84, 84n
Bangladesh 29, 138, 140, 182
banks 237, 239, 273 Cairns Group 122, 169
development 138, 139 Cambodia 158, 160, 161, 162, 163,
lending requirements 154–155 182
purchase of 152 Canada 23, 51, 98, 128, 129, 143,
secrecy 135 180–181, 182, 254
bankruptcy/restructuring proceeding Cancún Ministerial Meeting,
131 September 2003 58, 60, 62–63
bargaining process 8, 72, 74, 75, 82, 64, 85, 91, 92, 138, 158
85, 99, 102, 168–169 capacity-building 209
INDEX 299
capital 79, 115, 116, 138, 151, 164, Common Agricultural Policy (CAP)
169, 171, 174, 236, 238, 242, 59
272, 273 Commonwealth 216, 246, 251
access to 78 communications:
flows 17, 266 liberalization 245
goods 19 technologies 87
inflows 22 Community Reinvestment Act 79
loans 138 comparative advantage 25, 30, 52, 70,
markets 22, 31–32, 80, 124, 175, 89, 108, 116, 118, 120n, 149,
204 177, 194
imperfections 31–32; compensation 72, 153, 172, 173, 177,
liberalization 65, 80–81, 205, 213, 236, 266
150–151, 151n, 235, 251, 262, competition 4, 25, 27, 32, 69, 90, 104,
266; private 20, 31 110, 115, 119, 130, 148n,
mobility, free 150 164–165, 171, 194–195, 196,
rights 151 203, 216, 235, 262, 268, 269,
subsidies 231 275
Caribbean countries 63, 102, 248 see also anti-competitive effects
Caribbean Basin countries 137 incentive 133–134
Carlin, Wendy 237 law 19, 261, 268, 269
cartels 148, 263, 269, 270, 270f, 271 policies 53, 60, 65, 85, 147–148,
Central Africa 62 176, 247, 260, 261, 263,
Central European countries 196 267–273
certification assessment system 212 competitive advantage 184, 236, 269,
Chad 62 272–273
Chang, Ha-Joon 38, 52n, 78n, 89 computable general equilibrium
Charlton, Andrew 47, 48n, 119n, models (CGE) 199, 202–203,
133n, 134n, 266 215, 216, 227, 229, 232, 236,
Chile 19, 80, 151, 197, 246 243, 246, 249n, 255, 260
China 12, 14, 23, 38, 43, 45, 48, 52, Conceicao, Pedro 86n
63, 77, 97, 98, 126, 128, 129, concessions 116, 165
155, 158, 161, 169, 182, 196, conditionalities 208
220, 242, 246, 249, 255 political 165
Clarke, J. L. 269 conservatism 86, 149
clothing 46, 51 consumers 223, 230, 234, 255, 268,
collective action 86, 154, 262 270, 271
global 55, 74 benefits 258
Colombia 29, 197 goods 18–19, 45
Commander, Simon 248 prices 120, 123, 199, 217, 233, 237,
commercial presence see Mode 3 270
delivery service protection 212, 266, 272
commodities 19, 28, 45, 69, 109, sovereignty 86
121, 122, 123, 124, 146n, 173, consumption 121
184, 199, 209, 210–211, 217, abroad see Mode 2 delivery service
218, 218n, 220, 222, 233, tax 191
234n, 276 Convention on Biodiversity (CBD)
exports 211, 211f 146
prices 224, 233 coordination 65, 81
reforms 121–122 failures 29
subsidies 121n, 127 Corden, W. M. 28, 30
300 INDEX
core disciplines (WTO) 104 Deardorff, A. V. 52n, 117n, 243, 244,

corruption 22, 37, 134–135, 242, 267 255, 256, 257, 258
see also anti-corruption debt:
Costa Rica 197, 276 burden 53
cost-benefit analysis 267 cancellation 3
costs 28, 76, 82–83, 88, 90, 91, 111, Latin America 19–20, 21–22
123, 128, 130–131 153, 164, Decision on Measures in Favor of
171, 172, 181–182, 184, 192, Least Developed Countries
193, 233, 261, 262, 263, 272, (WTO) 204
275, 276, 277, 278 other costs; decision-making 38, 169
see also implementation costs; Dee, P. 117n, 243, 245, 246
labor, costs deliberations 167–168
environment 140 democracy 74, 235
legal 83–84 Desai, Mihir 248
production 154 development 56, 58, 63, 94, 97, 105,
reforms 172–173 141, 143, 155, 158, 173, 205,
Côte d’Ivoire 225 209–210, 272
Cotonou Agreement 2–3, 61n, 102n, aid 248–249
178 Doha Round 112t-114t
cotton: finance 55
prices 71 impact 68–73
production 61n, 62, 63 long-term 20
subsidies 59n, 71, 122 national 262
countervailing duties 52, 80, 81, policies 43, 60, 149, 261
121n, 127, 128, 154, 170, 217 restriction 65
Cranfield, J. 199, 200, 203 trade-related 208
credit 39, 53, 119, 124, 138, 152, 173, Development of Mauritius (DBM)
175, 203–204 204n, 213, 235, 210
237, 246 Development Round 3–5, 6–8,
constraints 230 41–42t, 42–56, 57, 67, 81, 86,
insurance 138–139 93, 126, 131, 141, 144n, 145,
trade 210 148, 150, 209 233, 252, 261,
crises see finance, crises 262, 277, 278
cross-border supply see Mode 1 Diamond-Mirrlees production
delivery service efficiency theorem 27
currency 273 Diao, Xinshen 50n
devaluation 155 Diaz-Bonilla, E. 50n
customs 193 differential effects 75–76
procedures 60, 261, 277 differential treatment 42, 42n
reforms 192, 193 Dimaranan, B. 50n, 121n, 122, 221,
valuation 48, 163 222f, 224, 229 230, 231, 232
cyclical industries 1301–131 diseases 55, 87, 163
Czech Republic 267n Dispute Settlement system (WTO)
76n, 82–83, 129, 170, 206,
Dasgupta, P. 26, 28, 32, 71n, 78n, 277
172n Dispute Settlement Understanding
de jure fairness 7, 76, 82–83, 130–131 (DSU) 83
de facto fairness 7, 76, 82–83, distortions 30–31, 39, 58, 59, 69, 71,
130–131 123, 124, 148, 154, 234
INDEX 301
distribution 56, 69, 74, 121 Easterly, William 172n

effects 122 Eastern countries 189
of income 28; Eastern European countries 196
Dixit, A. K. 27 Ebrill, L. 188, 189
Djankov, Simeon 269 economic analysis 5–6, 7, 68, 215
Doha Agenda 126, 193, 215, 262 economic development 3, 11, 13,
Doha Declaration 1–4, 55, 63, 73, 17–18, 22, 27, 33, 38, 43, 87,
160, 163, 217, 275n 89, 105, 120, 154, 173, 205,
Doha Ministerial Conference 57, 208
209, 252 see also development; industrial
Doha Round 3–8, 13, 41, 49, 52n, 53, development;
55, 57, 58, 60, 61, 64, 65, 73, 87, economic efficiency 25, 29, 71, 76,
91–94, 96–97, 104, 110, 118, 78, 79, 89, 115, 116, 119,
120, 144–145, 157, 163, 164, 150–151, 203n
172–173, 175, 176, 193, 196, economic growth 22, 23, 24–33,
197f, 205, 256, 258t, 261, 34–35, 35n, 73, 151, 152, 172,
262–263 177, 189–199, 203, 208, 214,
development issues 112t–114t 235, 237, 266
Doha Round Market Access see also free trade; endogenous
Proposal (MAP) 94–102 growth
Dollar, David 34, 195n, 198, East Asia 15, 17
198n Latin America 20, 22
domestic industries 6, 128, 274 poverty-reducing 215–216
domestic policy 4, 45, 246 trade, effects on 11, 13
domestic producers 78–79, 242–243, economic instability 54–55
246, 268; economic model 24–25, 124
see also foreign producers economic partnerships 102
domestic subsidies 123–124, 124n economic policies 14, 15, 17, 119,
see also subsidies 155, 235, 242
domestic support 58, 123, 124, economic power 74, 154, 269
233–234 economic principles 130–131
production 233 economic volatility 80–81
reduction 231 economics 12, 54, 78
reform 232t economy 16–17, 18, 26, 29, 37,
drugs 73, 136, 144 43, 52, 69–70, 79, 103, 126,
AIDS pandemic 62 138, 180, 197, 208, 213,
consumption 136 215, 222, 232, 244, 254, 269,
generic 61–62 273
dumping duties 52, 53, 65, 79, 126, distortions 191
127, 128, 130–131 131n, 147, domestic 133
170, 269 global 9, 88, 109, 115
industrial 90, 109
East Asia 12, 14–17, 22, 34, 45, 89, of scale 18, 25, 27, 29, 115
94, 103, 151 restructure 213
economic growth 15, 20, 89 system 139
export-oriented strategies 14–17 theory 153, 194, 197, 236
policies 13, 16, 20, 22 Ecuador 19, 77
government intervention 16 effective tariff rates 125;
East Asian Miracle 52 see also tariffs
302 INDEX
efficiency 227, 232, 236, 246, 258, exchange 115

275 manipulation 155
gains 227t, 245t, 251, 259t rates 19, 155
liberalization 228 expenditure 194, 273n, 274
wage theory 28n analysis 199
Egypt 95 government 273
Ellerman, David 249 export-oriented strategies 14, 46
emigration, temporary 248; exports 4, 6, 6n, 25–26, 34, 44, 45, 48,
see also migration 49, 50, 61, 62, 69, 94, 109, 118,
employment 26, 29, 30–31, 69, 108, 120, 121, 122, 124–126, 138,
109, 115, 120, 129, 175, 177, 139, 171, 173, 175, 177–178,
194, 203 179t, 181, 182, 184, 185, 187,
Emran, M. Shahe 28, 191 196, 199, 203, 205, 209–213,
‘Enabling Clause’ 100, 100n 219,222, 252, 255, 271;
endogenous growth 29; see also agriculture, exports;
see also economic growth; free manufacturing, exports
trade assistance 138–139
energy 140, 149 subsidies 58, 59, 103, 123, 123,
enforcement system 77, 85 125, 162, 199, 221, 223–224,
entrepreneurship 29, 39, 90 234
environment 140, 149, 151, 153, 167, elimination 233
266 supply 207
pro 135–137, 137n, 138 externalities 262–263, 275
standards 150, 168
equality 29, 54 factor 224
equity 56, 74, 77–78, 81, 110 endowments 194
Ernst and Whinney 276 payments 116, 116n
Estevadeordal, A. 181 prices 171, 171n, 199, 249
Ethiopia 19 factor-price equalization theorem
Europe 4–5, 11, 19, 77, 82, 109, 110, 115, 194
122, 123, 124, 126, 141, 147, farming 50, 218, 218f, 219, 224, 230
149, 178, 209, 218, 220, 227, incomes 233
233, 270 OECD 234
European Commission (EC) 58, 59, policies 224
83, 95n, 101–102, 127 129, 267, USA 58–59, 59n
276 finance 116–117, 138, 139, 170, 198,
European Union (EU) 2–3, 43, 44n, 206, 208, 213
51, 60–61, 62, 63, 64, 65, 92, 98, access to 209–210
124, 134, 137, 165, 169, 176, crises 138–139, 170, 198, 209–210
177–178, 180–181, 182, 187, development 3, 55
188, 199, 208, 210, 223, 233, financial markets 155, 194, 209
243, 254, 256, 266, 273, 275 liberalization 22, 53, 235, 237
preference dependency 185 financial services 116, 119, 152, 235,
Evenett, Simon J. 157n, 158, 161, 236, 241t
269, 273 liberalization 245, 246, 247
Everything But Arms (EBA) (EU) Findlay, C. 239
60–61, 61n, 165, 176, 177–178, Finger, J. M. 48, 48n, 60n, 170n, 192,
185, 186t, 187, 187t, 188, 208, 193
210 Fink, C. 148n, 247, 270
INDEX 303
Fischer, Stanley 34 general-equilibrium incidence

food 225, 227, 228, 229 analysis 68–69, 71, 109, 161,
prices 201, 224 174 178, 198, 224, 250, 276
processing 108, 111, 125 Generalized System of Preferences
protection 219t (GSP) 42, 97, 97n, 100
security 101, 127 101–102, 176, 178, 178n, 179t,
self-sufficiency 225 180, 182t
trade 220, 224, 226t rates 182–182
Food and Agriculture Organization generic drugs 61–62, 143, 145, 163
(FAO) 225 generic products 143–144
foreign direct investment (FDI) Ghana 19, 204
15, 22, 23, 104, 119, 134, Gibson, P. 183, 184, 185
134n, 150, 238, 238t, 242–243, global recession 1, 21
247, 249, 261, 264, 265, Global System of Trade Preferences
267, 275 among Developing Countries
see also Mode 3 service delivery (GSTP) 97
foreign exchange 19, 118–119 global warming 137–138, 140, 140n,
foreign producers 78–79, 246; 153
see also domestic producers globalization 74, 107n
François, J. F. 97, 117n, 121n, 168n, see also anti-globalization
195, 196, 196n, 197, 219, 224, economic 110
227, 228, 229, 229n, 231, 238, Goletti, Francesco 198
239, 244, 245, 256, 259t, 274 goods 87, 115;
free riding 45 see also industrial goods;
free trade 7, 12, 16, 20, 23–25, 30, manufacturing, goods
32–33, 33–34, 37 38–40, 89, domestic 251
102; industries 27
see also economic growth; labour-intensive 4
endogenous growth prices 199, 203
agreements (FTAS) 97, 101–102 Goulven, Katell Le 86
Freeman, R. 196 governments 104, 139
Friedman, J. 198 action 90
Fukasaku, Kichiro 211 failures 36–38, 39
Future of the WTO, The (Sutherland) intervention 16, 18, 20, 22, 24, 27,
100 29, 30, 31–32, 37, 89
loan programs 78
G20 63, 63n, 64 procurement 261, 263, 273–275
G33 64, 64n, 92, 92n access 274–275; liberalization
G77 209 275;
G90 64 policies 275, 275n
Gage, Jonathan 157n, 158, 161 revenues 27–28, 188–189
Gallup, John 109n Great Depression 18
Gappah, Petina 83 Green Box 124
General Agreement on Tariffs and subsidies 58, 59
Trade (GATT) 3n, 41–42, 42n, Green Room process 82, 169
43–45, 46, 67, 83, 84, 93, 93n, green tariffs 153;
100, 115–116, 118, 128, see also blue tariffs; tariffs
136–137, 237, 238, 239f, 247, Greenfield, J. 225
251, 261 Greenwald, Bruce 204
304 INDEX
Gropp, R. 188, 189 duties 183

Gross Domestic Product (GDP) 15, duty-free 210
52, 62, 75, 94–95, 96f, 109, 229, licensing 19
230, 234, 237, 243, 246, 250, threats 98–99
256, 269, 273, 274 incidence analysis 71, 71n, 75
per capita 95, 96f, 98n, 101, 109, see also general-equilibrium
117, 120, 195, 210 incidence analysis
Gross National Product (GNP) incomes 20, 28, 32, 33, 70, 180,
242–243 183, 194, 197, 198, 201–202,
Grynberg, Roman 157n, 161, 251 205, 217, 222, 226t, 227, 230,
Guinea 188 239, 242–243, 243, 248, 250,
251;
Hallwood-Driemeier, M. 150, 265 see also wages
Hamilton, C. 250 analysis 199, 201t
Hanslow, K. 117n, 243, 245, 246 distribution of 29, 171
Hausmann, Ricardo 29 household 224
health 151, 248 per capita 146
and safety 211 rural areas 199
Hertel, T. 50, 50n, 70n, 77n, 108n, taxes 28
117n, 121n, 122, 125, 126, 199, India 12, 19, 29, 38, 43, 48, 63, 97,
202, 203, 219, 221, 222t, 224, 100, 136, 220, 228, 248, 251,
227, 228, 229, 229n, 230, 231, 255
232, 243, 244, 245, 250, 252, indigenous knowledge 51;
253, 256, 257, 259 see also knowledge
Hoekman, Bernard 42, 90, 99, 104, individuals, movement see Mode 4
104n, 193, 224, 229, 234n, 239, delivery service
269, 269, 271 Indonesia 14, 19, 64, 198, 199, 200t,
Hong Kong 14 201, 201t, 202 220, 346
Horn, H. 83n industrial countries 3, 4, 7, 56, 73,
Hudec, R. E. 101 74, 75, 77, 80, 85–86, 88, 89, 93,
Hugo, Graeme 248 97, 98, 127, 136, 144, 153, 163,
human capital 30, 87 230, 232, 239, 242, 268, 271,
human rights 153–154 272, 273
Hungary 192 industrial development 12, 268
industrial goods 45–46, 65, 92, 94,
implementation costs 4, 60, 60n, 63, 108, 252
69–70, 71, 91–92, 94, 104, 149, liberalization 124–126
170, 171, 176, 191–193, 213, industrial policies 15, 16–17, 19, 22,
262, 264, 269, 276, 277 51–52, 90 98n, 103, 105, 120,
import substitution policies 14, 131, 263
17–22, 43, 45, 210–211 industrial products 48, 50
imports 6, 25, 43, 47, 50, 51, 60, 61n, industrialization 210–211
69, 71, 98, 100, 101, 102, 109, industries 87, 110, 164, 173, 175,
121, 123, 125, 127, 128, 129, 188, 203, 270
136, 138, 145, 171, 177–178, new 6, 27, 29, 210
182, 182t, 183, 189, 203, 213, protectionism 21
217, 220, 221, 222, 230, 233, inequality see wages, inequality
252, 254f(b), 255, 270, 270f inequities 78, 79
barriers 59, 98 trade regimes 80
INDEX 305
infant industry theory 20–21, 31, 79, protection 150–151, 265

91, 98; regimes 265–266
see also protection regulations 261, 264f
informal sector 71, 80, 191, 208 treaties 264–265
information 69, 74, 87, 119, 235, 236, bilateral 265f
242, 243, 246 investors 72n, 80–81, 104, 133–134
asymmetry 32, 82, 82n domestic 266
externalities 29 foreign 263, 264, 266
imperfections 31 protection 151–152, 236, 264
undisclosed 163 rights 150–151, 263
infrastructure 79, 171, 209, 210, 211, Islam, Roumeen 172n
212, 213, 273 Israel 129
institutions 205, 213, 267 Ivanic, M. 199, 202, 203
capacity 261 Iwakura missions 11
government 206
reforms 8, 167–170 Jamaica 225
insurance 203 Japan 11–12, 14–16, 51, 83, 98, 129,
markets 39, 70–71, 91 147, 154–155 169, 180–181,
Integrated Framework for 182, 227, 229, 242, 243, 254
Trade-related Technical Johannesburg Round 56
Assistance (IF) (WTO) 206–208 Johannesburg Summit on Sustainable
integration 34, 39–40, 99 Development, September 2002 3,
studies 208 55
intellectual property 32, 46, 48, 51, Jordan 129
61, 62, 73, 73n, 83, 85–86, 103, Joshi, Vijay 28, 29
107, 141–142, 142n, 143–145, Joy, R. M. 157n, 161
145n 146, 149, 162–163, 168, Jubilee 2000 campaign 3, 54
192, 262, 272 just-in-time inventory methods 276
interest rates 21, 22, 65, 81, 126, 131,
213, 273 Kangasniemi, Mari 248
International Monetary Fund (IMF) Kapur, Devesh 248
20, 33–34, 37, 65, 80, 81, 131, Kaul, Inge 86n
208, 241, 262, 266, 273 Keck, A. 93, 99, 100
international relations 55, 74 Kee, Hiau Looi 269
International Trade Organization Keen, M. 191
(ITO) 41 Keeney, R. 121n, 122, 221, 222, 230,
interventions: 231, 232
government 155 Kennedy Round 42n, 43–44, 44n, 45
regulatory 154 Kennett, Maxine 157n, 158, 161
investment 4, 22, 23, 50, 60, 73, 90, Kenney, R. 50n
103, 104, 119, 124, 131, Kenya 63, 225, 237
149–150, 171, 173, 177, 209, Ketkar, Suhas 117n
211, 216, 230, 236, 262, Kletzer, Lori G. 175n
263–264, 264f, 265–267 knowledge 30–31, 32, 109, 142, 143n,
adjustment-related 204 145, 146, 236, 242, 243, 246,
agreements 149–152, 263–264, 272
265–266 North-South gap 51, 143
competition 134, 267 Konandreas, P. 225
foreign 265, 266 Kouaouci, Ali 248
306 INDEX
Korea 12, 15 157, 160, 161, 161n, 162, 163,

Kose, M. Ayhan 80n 176, 177–179, 180–181, 181t,
Kost eckki, Michael M. 42 182, 183, 184–185, 201, 203,
Kraay, A. 195n, 198, 198n 206, 208–213, 248
Krueger, Anne 37 accession negotiations (WTO)
Kuznets, S. 14, 15 160n
Kyoto Protocol 137, 153 trade preferences 176–188
Levenstein, Margaret C. 270
La Porta, Rafael 269 Levine, R. 237
labor 6, 8, 26, 115, 116, 149, 151n, Levinsohn, J. 198
153–154, 199, 202, 236, 238, Lewis, W. A. 14, 18
251 liberalization 1–2, 6–7, 9, 12, 13, 17,
see also migration; workers 23–35 35–36, 38, 39, 41, 43, 44,
child 154 47, 49, 50, 52, 57, 58, 65, 69–70,
conditions 150 71, 72, 72n, 73, 78, 80, 81, 82n,
costs 154, 174 89, 90, 93–94, 95, 97–99, 102,
flows 275 103, 104, 107n, 108–110, 111,
markets 194–203, 216, 247, 260 115, 121, 127, 129, 152, 170,
mobility 111, 115–120, 150–151, 172, 172n, 173–174, 175–176,
250 180, 185, 186t, 189, 190,
rights 151 193–196, 197, 197f, 198, 198f,
unskilled 109, 152, 194, 197, 249, 199, 200t, 201, 201f, 202, 202t,
252 203, 204, 205, 208–214,
labor-intensive goods 93 215–219, 222–233, 236, 255n,
labor-intensive services 52, 52n, 260, 261, 268–269
115–120, 152, 247 see also adjustment costs,
Lacarte-Muro, Julio 83 liberalization; agriculture,
Laird, S. 46n, 178, 182 liberalization; capital markets,
laissez-faire economy 16, 89 liberalization; financial
Lall, S. 89 markets, liberalization;
land: financial services,
price 172 liberalization; food,
seizure 154 liberalization; industrial goods,
Laos 158 liberalization; manufacturing,
Latin America 13, 19–22, 94, 123, liberalization; poverty,
151, 220, 228, 233, 248, 269 liberalization; services,
see also import substitution liberalization
policies global 259f
learning 30, 31 instruments 123
Lederman, D. 23 multilateral 8, 177
legal disputes 76–77, 83; national circumstances 13–14
see also Dispute Settlement system preferential 109
(WTO) revenues 191t
assistance 170, 206 services 53, 117
lending requirements 79, 152 South-South 96
banks 154–155 tariffs 184, 185, 185t
Lesotho 237 licensing, compulsory 143, 144, 144n
Less Developed Countries (LDCs) Lightfoot, J. E. 191
60–61, 83, 84, 87, 95, 98, 99, liquidity 116–117
INDEX 307
local industries 25–26 failures 27, 30, 31, 36, 37, 39, 89
Lopez-de-Silanes, Florencio 269 foreign 25, 171, 247
Love, James 143, 143n, 144 free 268–269
low-income food-deficit country imperfections 31, 109, 194–195,
(LIFDC) 225 196, 197
Low, P. 93, 99, 100 international 227
Lucenti, Krista 275 openness 12, 34, 88, 90, 99, 108,
lumber 154 109, 115–131, 139, 198f
Luxembourg reform (2003) 59 prices 200t, 233–234
Lyakurwa, B. 168n support 123
share 196, 271–272
macroeconomics 15, 34, 88n, 213, with OECD 222t
274 Marrakesh Agreement 158
Malawi 187, 225, 237 Martin, Philip 249
Malaysia 14, 15, 136, 210–211, 211f, Martin, W. 46n, 49, 50, 77n, 117n,
246, 256 121n, 125, 126, 196n, 219, 224,
Mali 62 227, 228, 229, 229n, 244, 245,
Mann, C. 276 256, 259
manufacturing 94, 110, 125, 182, Martinez, Giralt, Xavier 269n
201, 216, 234, 240, 241, 242, Massey, Douglas S. 248
243, 244, 246, 252–259, 275 Mattoo, A. 118, 148n, 241, 246, 247,
exports 252, 253f(a), 253f(b) 270
goods 50, 52, 77n, 79, 108, 109, Mauritius 210
117, 126, 199, 201 Mavroidis, P. 83n, 193, 269, 271
labor-intensive 8, 110–111 Mayer, Colin 237
liberalization 126, 246t, 255, 258t, McAfee, R. P. 274
259t McDougall, R. 222
tariffs 218, 255t, 256, 256f, 257, McGuire, G. 239
257t, 258 McHale, John 248
peaks 260 McMillan, J. 274
markets 2, 38, 39, 43, 44, 45, 69, 72, medicines 145, 163
87, 89, 153, 209, 211 219, 269, traditional 146
272 Meier-Ewert, M. 138
see also financial markets; labor, Mellinger, Andrew 109n
markets; risk, markets; non- Melo, Jaime de 174
market access; preferential membership (WTO) 159
market access criteria 160
access 3, 31, 47, 58, 60, 63, 65, Menoza, Ronald U. 86n
76, 88, 89, 94, 97, 98, 99, 102, Menendez, A. 23
108, 110, 118, 120, 160, 175, mercantilism 67
177, 184–185, 191, 194, 204, merchandise trade 94, 117
208, 209, 215–218, 220, 230, metrology 212
238, 239f, 247, 260, 261, Mexico 19, 22–24, 45, 63, 123, 129,
276–277 192, 197, 220, 233, 246
free 94–95, 98, 108–109; Michalopoulos, C. 90, 99, 170,
preferences 180; rules 176 206
distortions 222 micro-enterprise 116
domestic 164–165, 210 micro-finance 116
economy 26–27 Middle East 220, 232
308 INDEX
migration 8, 107, 248, 250, Nelson, D. 274

250n, 252 neo-classical models 6, 69, 70
illegal 252 neo-liberalism 20, 88–89, 91
liberalization 251 Nepal 158, 160, 161, 162, 163
temporary 111, 230, 247–252 net effects 121, 122
Millenium Development Goals New Zealand 129, 220
(MDGs) 3, 3n, 55 Newbery, David M. G. 27, 71n,
minerals 135, 154 195
Minot, Nicholas 198 Ng, F. 234n
Mode 1 service delivery 118, 118n, Nigeria 19, 83
237, 239 Noland, M. 15, 89
Mode 2 service delivery 118, 118n, non-agricultural businesses 199,
119, 237 200, 202
Mode 3 service delivery 117, 117n, non-agricultural goods 57;
118, 119, 238 see also goods; industrial goods
Mode 4 service delivery 115–117, non-agricultural products 252
216, 230, 238, 239, 247–248, non-bribery legislation 134–135
251 non-competitive markets 76;
monetary policies 194, 208 see also markets, competitive
monopolies 268, 270 non-core disciplines (WTO) 104,
power 141, 144, 272 104n
rents 163 non-industrial countries (NICs)
monopolization 269 180;
Monterrey Round 56 see also industrial countries
Morocco 190 non-market access 118, 260;
Most-Favoured Nation (MFN) 45, 61, see also markets, access
93, 93n, 97, 99, 102, 159, 161, non-reciprocal preferences 60, 100,
164, 175, 176, 177, 178, 179t, 181t
180t, 182t, 188 see also reciprocity
tariff rates 182–183, 184–185, non-tariff barriers 8, 24, 45–46, 49,
185n, 255, 256f 52, 65, 79, 107, 111, 126–131,
reduction 187 189, 252, 255
Mozambique 175 see also tariffs
Multifiber Arrangement (MFA) 44 non-trade distortions 124, 169
multilateral action 266, 271 see also distortions
Multilateral Agreement on Norman, V. 27
Investment (MAI) (OECD) 264 North Africa 123, 220, 232, 233
Multilateral Investment Guarantee North America 12, 122, 233
Association (MIGA) 265–266 North American Free Trade
multilateralism 225–226, 262 Agreement (NAFTA) 22–24,
agreements 264 143, 151, 167, 197, 266
Mutti, John 174 North-South knowledge gap 51,
143
narcotics 135 Norway 129
national treatment 129–130, nutrition 230
273–274
natural persons, movement see Mode objective access criteria 101
4 service delivery offshore outsourcing 118
natural resources 89–90, 109 oilseeds 123
INDEX 309
Olarreaga, M. 234n, 269 Poland 128, 193, 277

Oman, Charles 133n policies 1–2, 13, 14, 16, 17, 19, 22,
Oostendorp, R. 196 31–32, 34, 35, 79, 89, 90, 173,
Organization for Economic 174, 209, 213, 273;
Cooperation and see also domestic policy; economic
Development (OECD) 46, policies; industrial policies;
46n, 47–48, 50, 51n, 53, 58, taxes, policies
59, 60, 93, 108n, 119n, 120, complementary 203–208
120n, 122, 123, 124, 125, coordination 267
133n, 150, 176, 193, 196n, national 274
216, 218, 220, 224, 225, space 85–86, 103
227, 229, 230, 231, 231t, political conditionality 165
233–234, 234n, 238, 241, political economy 72, 241–242
252, 255, 264, 265, 269, 271, pollution 168
273, 274 poverty 3n, 6, 13, 15, 23, 24,
Organization of Petroleum Exporting 54–55, 70, 73, 151, 194, 201,
Countries (OPEC) 271 201n, 202, 202t, 203, 217,
Oswald, Lynda J. 270 218
Other South-East Asia 255 economic growth 215–216
Otsuki, T. 276 liberalization 224, 225, 237
outcome fairness 82; reduction 237
see also procedural fairness Poverty Reduction Strategy Papers
output 26, 30–31, 99, 124, 125, (PRSPs) 208
226–227, 231, 242, 250, 264 power 86
outsourced goods 120 asymmetry 82, 108
Oyejide, A. 168n relationships 43, 75
Prasad, Eswar 80n
Pacific states 63, 102 Preckel, P. 199, 202, 203
Pack, H. 15, 89 predatory pricing 130, 262;
Pakistan 19, 29, 136, 138 see also prices
Palmeter, N. D. 274 preferences 100, 101, 102, 176–180,
Pareto-efficiency 26, 28, 29 181n, 183, 187, 274
Pareto-superior 148n elimination 184
patents 145, 146, 146n, 163 erosion 184–185, 188
laws 51, 51n, 143–144 margins 188, 213, 274
presumptions 146 non-reciprocal 184
payments 123–124, 233–234 rates 185
land-based 231 systems 178
made to foreign companies 135 under-utilization 181–182
Pellegrino, Adela 248 preferential access 48, 61, 98, 181,
penalties 77 188
Perry, G. 13 preferential markets 226–227
Peru 19, 185, 202, 225 access 88, 95, 97, 165, 176
personnel 247; preferential schemes 99–102,
see also labor 181
pharmaceutical industry 85–86, preferential treatment 72, 95, 99
163 price-fixing conspiracy 270
Philippines 64, 256 price-sensitive necessities 123,
phytosanitory conditions 52, 170 233
310 INDEX
prices 26–27, 123–124, 125, 145, 196, standards 209, 211–213

199, 222–223, 226–227, surplus 208–209
230–231, 231t, 232, 235, 237, temperate 220
270, 274 progressivity 77, 97–98
see also consumers, prices; factor, protection 32, 34, 44, 47, 49, 50n, 65,
prices; food, prices; goods, 72n, 73, 77, 78, 79, 85, 88,
prices; markets, prices 98–99, 110–111, 121, 126, 139,
agriculture 71, 120 199, 211, 219, 227, 233–234,
cotton 71 238–239, 240t, 255, 261
elasticity 122 see also agriculture, protection;
redatory 79n infant industry theory;
relative 71, 75, 75n, 171, 199, 202, investment, protection;
204, 223 subsidies
rigid factor 29 cuts 225
sensitivity 121 elimination 218
services 119 mechanism 230
shocks 225 reinstrumentation 123
uncertainty 71 protests see anti-globalization
volatility 90–91 movement
welfare, impact 224–225 provisions 81, 88
privatization 131, 235, 268–269 redistributive 72
pro-development 8, 69, 110, 152 Prowse, S. 206
see also development public goods 39, 86n, 153
procedural deficiencies 168 public health 51, 62, 143, 144, 163,
procedural fairness 81–82; 163n
see also ‘outcome fairness’ public revenue 188, 190–191
processed foods 51, 122, 125, 173, punishment mechanisms 84
254
processed goods 108 Quad countries 98, 100, 181t
processed products 125 quotas 71, 71n, 76, 123, 172, 174,
procurement see government, 223, 233
procurement reductions 191–192
production 6, 59, 76, 80, 87, 90–91,
99, 115, 123, 124, 128, 133, Rae, Allan N. 234n
142, 154, 171, 205, 209, 209, Raghavan, C. 49n
212, 226–227, 233, 234, 241; Rama, M. 195, 195n, 196, 198, 198n
see also cotton, production Rasul, I. 204
benefits 258 Ratha, Dilip 117n
low-income 210 Rathindran, R. 148n, 241, 247, 270
subsidies 218, 223–224, 230, Raven, John 276
231–232, 234 Rawls, John 81, 81n
production-based support 124 recession 274
production efficiency theorem see reciprocity 88, 97, 109
Diamond-Mirrlees see also non-reciprocal preferences
production efficiency theorem co-ordinated 246–247
productivity 16–17, 26, 32, 70, 116, redistributive effects 120, 121, 173
120n, 195, 230, 250 reforms 9, 34, 35–36, 38, 39, 40, 44,
products 121, 177–178, 181 47, 50, 65, 69, 76, 79, 83n, 88,
marginal 116, 116n 90, 91, 116, 117–118, 123, 129,
INDEX 311
147–148, 151, 157, 171, 172, markets 6, 8, 26, 71, 99, 115,
173, 174, 187, 189, 191, 192, 194–195, 197, 216
193, 194, 196, 203, 205, premium 80–81
213–214, 215–218, 241, 243, Robinson, S. 50n
246, 252, 257–258, 260, 262, Rodrik, Dani 29, 30, 34, 38, 46n, 88n,
269, 277 90, 195
see also agriculture, reforms; Rodriguez, Francisco 34, 88n, 195
commodities, reforms; Rogoff, Kenneth 80n
institutions, reforms; services, Roland-Holst, D. 224
reforms Rosenstein-Rodan, P. 18
costs 172–173 ‘Round for Free’ 92, 92n, 93–94, 98
expenditure 204, 204f royalty payments 103
multilateral 233, 261 rules 133–134, 173–174, 176, 193
regional trade agreements 164–165, of law 86, 130
169–170 of origin 95–96, 109, 181, 181n,
regulations 74, 173, 193, 206, 211, 182
240, 242, 264, 266, 276; of reason 272
see also Singapore Issues ‘rules-based’ trading system 87, 94
agreements 120, 176, 192 Russia 271
disciplines, multilateral 60
domestic 261 Sachs, Jeffrey D. 34, 88n, 109n, 195
environment 236 Safadi, R. 46n, 178n, 182
elimination 235 safeguard measures 52, 128–129, 170
harmonization 261, 262 Samoa 159
technical 211–212 Samuelson, Paul A. 25
Reinhardt, Eric 83, 84, 84n Samuelson-Stolper theorem 197
remittances 116, 116n, 117, 117n, Sachs, Jeffrey D. 34
225, 248–249 251, 252 safety 151
rents 194, 242–243, 246, 255, 268 safeguard measures 127, 128–129
rent-seeking 37 safety nets see social safety nets
Report of the High-Level Panel on Sala-i-Martin, Xavier 195
Financing for Development Samoa 158
(Zedillo Commission) 91 Samuelson, Paul A. 171n
representation 168–170 sanctions 77, 154, 154n
representativeness 169 savings 14, 203
research 73, 142, 146, 272 Schuler, P. 48, 170n, 193
finance 145 Sen, A. 81n
Research & Development Senegal 189
(R & D) 30 sequencing 26, 31, 38
resources 6, 20, 25–26, 28, 69–70, 88, service delivery see Mode 4 service
192, 205, 213 delivery
allocation 230, 243 services 45, 46, 52, 65, 65, 87, 107,
reallocation 222 109, 111, 115, 118, 205, 216
restrictions 12, 25, 27, 34–35 contracting out 236
reverse engineering 103 liberalization 117–120, 234–244,
Ricardo, David 12, 25 244t, 245, 245t, 246–247, 275
Richardson, J. David 174 openness 241f
risk 199 reforms 117, 118, 120
exposure 70–71, 197 Shapiro, Carl 175n
312 INDEX
Sharma, R. 225 Stevens, C. 101

Shleifer, Andrei 269 Stiglitz, Joseph E. 23, 26, 27, 28, 30,
Shrimp-Turtle case 136–137, 153, 32, 47n, 52n 71n, 78n, 86n,
153n 89n, 128n, 151n, 172n, 175n,
Singapore 14, 15, 80, 151 191 195, 204
Singapore Issues 4, 53, 59–60, 60n, Stolper, Wolfgang 171n
63, 141, 173, 176, 191–192, Stotsky, J. 188, 189
193, 261–278 Straubhaar, Thomas 249
Singapore Ministerial Meeting, 1996 structures (WTO) 168–170, 268
(WTO) 4, 53, 59n, 206 Strutt, Anna 234n
small and medium-sized domestic sub-Saharan Africa 47, 123, 220, 228,
enterprises (SMEs) 152, 232, 233, 238, 257
154–155, 235, 246, 267, 273, subcontracting out 251, 275
275 Subramanian, A. 241
Smith, Adam (1723–1790) 12, 115 subsidies 12–13, 28, 29, 30–31, 32,
social insurance 26, 99, 102 59, 76, 78, 88, 122 131, 134,
social justice 5–6, 74, 110 139, 219, 221, 222, 229, 233,
social policies 22, 119, 120, 235 234, 267, 269
social safety nets 8, 26, 39, 71, 129, see also agriculture, subsidies;
203–204, 213 cotton, subsidies; domestic
social security, expenditure 204, 204f subsidies; exports, subsidies;
Solomon Islands 188 production, subsidies
South Africa 63, 235n, 251 domestic 123–124
South Asia 178, 220, 251, 251 production 21
Rest of 256 elimination 122, 123, 189, 223
South-East Asia 210–211 entrepreneurship 90
South Korea 14, 30, 129, 138 production-limiting 59
South-North trade 253–254 reduction 63
South-South trade 94, 97, 102, 126, technology 105
164–165, 216, 252 255–259 under WTO rules 58
liberalization 96–97, 164 Sudan 158
special and differential treatment supply constraints 209–213
(SDT) 88, 90, 91–94, 95, 96, 99, ‘surrogate country methodology’ 128
100, 101, 105, 162, 217 Suslow, Valerie Y. 270
special interests 242 sustainable development 136
politics 126 Sutherland, P. 100
spillovers 16–17, 30, 103, 119, 249, Sweden 276
262, 271 Switzerland 129
Srinivasan, T. N. 32, 33, 46n
standards 153 Taiwan 12, 14
national body 212 Takacs, Wendy E. 174
products 209 Tanzania 210–211, 211f
technical 211–212 tariffication 49–50, 71, 189
standards of living 43, 70, 122, 233 tariff-rate quotas 47, 92
state aid guidelines (EU) 134, 267 tariffs 2, 23, 27–28, 30, 35n, 38,
steel 129, 130, 174 41–42, 43, 45–46, 47, 49, 50, 51,
Stephens, Malcolm 138 57, 60, 61, 71n, 75n, 77, 84,
Stern, R. 52n, 117n, 243, 244, 255, 84n, 90, 94, 97, 99, 102, 108,
256, 257, 258 108n, 123, 124–125, 126,
INDEX 313
127–128, 129, 130, 160, 164, Western 11

171, 173, 174, 175, 176, telecommunications 235, 241
177–178, 179t, 181t, 182, 184, terms of trade 25, 70, 72, 75, 76, 122,
187, 188–189, 217, 219, 230, 222, 227, 228, 230, 231, 232,
252, 254, 256; 233, 255
see also agriculture, tariffs; textiles 2, 44, 46, 47, 47n, 51, 108,
liberalization, tariffs; Most- 111, 125, 173, 174
Favoured Nation (MFN), tariff Thailand 14, 136, 202
rates; liberalization, tariffs Thomas, M. 50n
averages 182t Tokyo Round 42n, 44, 45
barriers 8, 111, 125n, 218 total factor productivity 16
binding 161, 161f, 162n, 162 sectoral 16–17
cuts 201, 233 tourism 118
elimination 189, 233 toxic wastes 151
peaks 50–51, 125, 254, 255 trade 263
USA 183 agreements 68, 71–72, 78–79, 81,
rates 183, 183f, 199–200 81n, 86, 105, 107, 107n,
reduction 3–4, 12, 13, 41, 50, 60, 108–114, 126, 133, 134, 139,
77, 79, 88, 92, 102, 108, 122, 140, 146, 149, 153, 164, 184,
161, 172, 175, 184n, 190, 194, 208, 214, 233, 241–242
191–192, 208, 209, 252, bilateral 80; exemptions 88;
255–256, 277 procedures 167–168
revenues 125, 172, 175–176, barriers 52, 72–73, 117, 222, 235,
188–191, 190t, 208 239–240, 242–244 246
of GDP 188, 189, 189f removal 229t, 245–246
Tarr, David 174 capacity 206, 208
taxation 225 changes 198f
system 190–191 world 228t
taxes 28, 68, 71, 125, 133, 133n, 134, disputes 82–83, 86, 136
150, 172, 175–176, 188, diversions 170, 176–177
190–191; facilitation 261, 263, 275, 275n,
see also tariffs, taxes 276–278
corporate, reduction 267 negotiations 1, 2, 3, 4, 5–7, 8, 36,
policies 65, 71, 80, 81 40, 43, 45–46, 54 63, 67, 74–75,
practices, harmful 119n 78, 79–90, 93, 100, 101, 102,
revenues 23, 27, 189, 191, 208, 213 107, 108, 109, 110, 110n, 141,
Taylor, J. Edward 248 151, 168, 169, 216, 217, 235,
technical assistance 170, 204–208, 246–247, 252, 262, 266, 272
209, 212, 262, 263, 276, 277 openness 195
Technical Barriers to Trade performance 205
Agreement (TBT) 162 policies 30–31, 38, 39–40, 45, 88,
211–212, 212n 135–136, 137, 153, 159, 168,
technologies 78, 105, 144, 146n, 171, 205, 206
209, 212 preferences 97n
adoption 196 procedures 261, 277
gap 61 regimes 35n, 28, 39, 71, 74, 159,
policy 103 175, 213
transfer 144–145, 163, 197–198, distortions 221; global 110n;
202 inequities 80
314 INDEX
trade (cont.) 224, 236, 252, 256, 257t, 262,

specialization indices 221t 273
support 207 US African Growth and Opportunity
surpluses 155 Act (AGOA) 3
taxes 27–28, 213 US Farm Bill 58–59
world 254f(a) US Federal Reserve 21
Trade-Related Aspects of Intellectual USA 1, 4–5, 11, 19, 22–24, 43, 45,
Property 50–51, 54, 57, 58 59, 60, 61,
Rights (TRIPS) 45–46, 49, 51, 61, 62, 63, 64, 65, 71, 77, 79, 80,
93, 103, 140, 143, 144, 82, 83, 98, 100, 105, 109, 110,
145–146, 149, 159, 163, 163n, 123, 124, 125, 126, 127, 128,
167, 192, 262, 272 128–129, 131, 134–135,
trade-related assistance 207t 136–137, 138, 141, 143–144,
Trade-Related Investment Measures 145, 147, 149, 151, 152, 153,
(TRIMS) 45–46, 49, 103, 149 153n, 154–155, 159, 164, 165,
tradeable goods 27, 149 169, 174, 175, 176, 180–181,
trading opportunities 12, 14, 39, 88, 183, 184, 185, 185t, 199, 218,
209 219, 220, 223, 227, 233, 234,
trading system 43, 94, 169, 177 235, 236, 239, 242, 243, 248,
multilateral 73, 103, 164 254, 262, 270, 271, 272, 273,
training 30–31 275, 276, 277
transparency 82n, 168, 273–274 utilities 119n
negotiations 8, 82
working group 4 value-added tax (VAT) 28
tribunal, international 130, 146 van der Mensbrugghe, D. 224
tropical products 122, 123, 233 van Meijl, H. 97, 117n, 195, 196,
trust funds 206 197
Tunisia 193, 277 van Tongeren, F. 97, 117n, 195, 196,
Turrini, A. 178, 182 197
Vandenbussche, Hylke 269
Uganda 95, 237 Vanuatu 158, 160, 162, 163
unemployment 8, 25, 26, 35, 39, 70, Venezuela 202
70n, 71, 91, 99, 102, 171, 175, Verikios, G. 117n, 245
194, 199, 203, 213, 225, 230 ; Vietnam 159, 198–199
see also employment voices 82, 82n, 93, 169
United Kingdom (UK) 44, 64, 248
Industrial Revolution 11 wages 4, 23, 28–29, 31, 153, 154, 194,
United Nations (UN) 87, 145–146 195, 196f 196, 197–198, 199,
United Nations Millenium Summit, 241, 250, 251;
New York September, 2000 3, see also incomes
55, 56 data 196–197
Uruguay 197 dispersion 198f
Uruguay Agreement on manufactures inequality 6, 14–15, 23, 28, 54–55,
255 71, 196–197 198f, 198n
Uruguay Round 2, 44n, 45–46, 46t, unskilled 197f, 198
47, 47n, 48, 48n, 49, 51, 52, 53, Wainio, J. 183, 184, 185
60, 60n, 61, 76, 79, 82, 91–92 Walmsley, T. 250, 251
103, 118, 126, 149, 176, Wang, Z. 250
191–192, 193, 211, 212, 219, Warner, Andrew 34, 88n, 89, 195
INDEX 315
Washington Consensus 2, 2n, 17, 22, less skilled 247

88–89 mobility 70
Wei, Shang-Jin 80n skilled 248, 251
welfare 71, 151, 173, 178, 199, 205, unskilled 111, 150–151, 249
215, 216, 225, 230, 231, 232t, urban 233
250, 251, 255, 256, 258, 260, World Bank 20, 46, 51, 61, 89, 94,
271 109n, 170, 193, 197 204, 205,
economic 237 205n, 241, 254, 277
effects 121, 174, 180, 180t, 217, World Social Forum 3, 54
219, 229, 244t, 269 World Trade Organization (WTO) 1,
enhancement 6, 24–33, 191 3n, 4, 7, 8, 38, 40, 41, 48 49, 51,
expenditure 204, 204f 53, 53n, 55, 59–60, 60n, 62–63,
gains 2, 24, 26, 29, 46, 55, 73, 94, 65, 68, 75, 77, 79, 81, 82n, 83,
97, 117, 120, 126, 205, 216, 85, 86, 87, 88, 88n, 94, 95, 96f,
227t, 228, 229, 229t, 230, 234, 97, 98, 99, 101, 102, 103, 104,
234n, 243, 243n, 245t, 246, 108, 117, 118n, 119, 120, 121,
247, 255–256, 257t, 258t, 259t 124, 128, 130, 134, 136, 137,
losses 121, 174, 250 139, 142–143, 144, 146, 150,
prices, impact 224–225 157, 158–160, 162, 169, 170,
reducing 28, 38–39 173, 175, 192, 193, 196, 205,
systems 71 209, 260, 261, 262, 263, 264,
trade-off 69 265, 268, 271, 272, 276, 277
West Africa 62 see also accession negotiations
Western Europe 227–228, 229, 232, (WTO); Dispute Settlement
269 System (WTO)
Western powers 15 agreements 90, 192, 206
Whalley, J. 250 Goods Council 4, 59n
Wilson, J. 276 obligations 99–102
Wilson, John S. 277 rules 75, 84, 89, 128, 129, 136n,
Winters, A. 33, 36, 37 137, 143–144, 159, 164
Winters, L. A. 46n, 49, 90, 99, 168n, world trading system 1, 7, 17, 38,
174, 248 250, 251 44–45, 49, 55
women, discrimination against 55 WTO-plus conditions 159
workers 194, 196–197, 197n, 203 Wunsch, S. 118
see also employment; labor;
migration; unemployment Zambia 19, 202
bargaining 154 Zedillo Commission 91
disadvantaged 155 Zhang, Xiao-guang 117n, 245, 251
foreign 249, 251 Zoellick, Robert 1, 5n, 57, 64, 65, 65n
laid-off 171, 174, 175, 194 Zwart, Tony 225

(Initiative For Policy Dialogue Series C) Joseph E. Stiglitz, Andrew Charlton-Fair Trade For All - How Trade Can Promote Development-Oxford University Press (2006) PDF

Загружено:

Сведения о документе

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

(Initiative For Policy Dialogue Series C) Joseph E. Stiglitz, Andrew Charlton-Fair Trade For All - How Trade Can Promote Development-Oxford University Press (2006) PDF

Загружено:

Авторское право:

Доступные форматы

Fair Trade For All

This page intentionally left blank

Joseph E. Stiglitz and Andrew Charlton

ISBN 0–19–929090–3 978–0–19–929090–1

At the turn of the millennium, there was a new sense of collective

demanding that we sign a fair agreement. If we return with another

only marginally related to constraints on external market access,

should be provided by the advanced industrial countries?’ But before

Part of the problem has been that discussions on reforming the

1 Introduction: The Story so Far 1

2 Trade Can be Good for Development 11

3 The Need for a Development Round 41

4 What has Doha Achieved? 57

5 Founding Principles: The Basis of a Fair Agreement 67

6 Special Treatment for Developing Countries 87

7 Priorities for a Development Round 107

8 How to Open up Markets 115

9 Priorities Behind the Border 133

10 What should not be on the Agenda? 141

12 Institutional Reforms 167

13 Trade Liberalization and the Costs of Adjustment 171

Appendix 1: Empirical review of market access issues 215

A1.3 Trade specialization indices, 1965–1998 221

A1.9 Average MFN tariff on manufactures, 1995 and 2005 256

Agreement on Agriculture WTO agreement which focuses on improving

anti-dumping duties Speciﬁc import duties imposed by importing countries

anti-globalization A political stance of opposition to the perceived negative

Cairns Group A group of countries which lobby together for agricultural

competition policy Policies designed to protect and stimulate competition

compulsory license Authorization for a government or company to make

Cotonou Agreement A treaty signed in Cotonou, Benin, in June 2000

countries on foreign aid, trade, investment, human rights, and governance;

countervailing duty A means to restrict international trade in cases

development box Measures proposed to give developing countries special

Dispute Settlement Body The General Council of the WTO, composed

Doha Declaration Statement made at the fourth WTO ministerial conference

externality A side-effect or consequence (of an industrial or commercial

G33 A group actually consisting of 42 developing countries of the WTO.

General Agreement on Tariffs and Trade (GATT) An organization established

General Agreement on Trade in Services (GATS) WTO agreement concluded

Generalized System of Preferences (GSP) A program to grant trade advantages

government procurement Purchase of goods and services by governments

Green Room Closed meetings during which developed countries negotiated

import quota A form of protectionism used to restrict the import of goods

infant industry protection Protection of a newly established domestic

Jubilee 2000 An international coalition which called for cancellation of

market access The extent to which a country permits imports. A variety of

market failure A case in which a market fails efﬁciently to provide or

Mercosur A trading zone consisting of Brazil, Argentina, Uruguay, and

Millennium Development Goals (MDGs) Goals which governments

mode of supply WTO term to identify how a service is provided by a

most-favored-nation (MFN) treatment A country extending to another

national treatment Treating foreign producers and sellers the same as

necessity test Procedure to determine whether a policy restricting trade is

non-tariff barriers (NTBs) A catch-all phrase describing barriers to interna-

Organization for Economic Co-operation and Development (OECD) Group

Pareto efﬁciency The criterion which stipulates that for change in an

patent A grant from a government to a ﬁrm conferring the exclusive

Poverty Reduction Strategy Paper (PRSP) A document describing a country’s

production subsidy A payment perhaps implicit, by government, to producers

protocol of accession Legal document recording the conditions and obliga-

Quad countries Canada, the EU, Japan, and the US.

quota Measure restricting the quantity of a good imported or exported.